Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Implementation of an algorithm for tapering analgosedation reduces iatrogenic withdrawal syndrome in pediatric intensive care

Implementation of an algorithm for tapering analgosedation reduces iatrogenic withdrawal syndrome... Editorial CommentThe study proposes a simple algorithm to taper analgosedation in a pediatric intensive care setting to reduce iatrogenic withdrawal syndrome (IWS) in children requiring sedation for more than 5 days. The use of peak and sum WAT scores is a useful indicator of the severity and prevalence of IWS. The algorithm still requires validation in different pediatric intensive care settings.INTRODUCTIONProper analgosedation is essential in the treatment of critically ill mechanically ventilated patients in pediatric intensive care units (PICUs). Analgosedation where pain is addressed before sedation is essential for safe and respectful care.1 The most common agents used for the past decades are opioids such as fentanyl and morphine, and sedatives such as midazolam.2 Pediatric intensive care patients need continuous infusions of these medications to maintain an adequate level of comfort. Further, intensive care nurses must prevent unplanned extubation and loss of vascular accesses. However, the use of these medications over time may lead to side effects such as iatrogenic withdrawal syndrome (IWS), prolonged mechanical ventilation (MV), and increased PICU length of stay (LOS).3 If analgesics or sedatives are tapered too rapidly, the patient is at risk of IWS, which is determined by a typical pattern of signs and symptoms showing central nervous system drug effects.1 The symptoms cause discomfort during the tapering phase. In an earlier study, we found that tachycardia, agitation, sleep disturbance, crying, and tremors were the most frequent signs and symptoms in the current study units and that they prompted additional bolus medication.4 IWS can result in continued and adverse health consequences for the patients, including seizures, myocardial ischemia, and increased length of PICU stay and hospital stay.5,6Different strategies have been described to reduce the side effects of IWS, such as following sedation protocols7 and performing daily sedation interruption.8 Nonetheless, assessment, prevention, and management of IWS continue to be challenging.9 Best et al. identified risk factors associated with IWS and found that duration of therapy and cumulative doses of opioids and benzodiazepines are the most predictive factor of IWS.10 They also highlighted the need to further explore process, and system variables, such as sedation/IWS assessment, and protocol adherence.10 The incidence of IWS in recovery is highly variable in different studies, ranging between 47% and 95%,4,11–13 and for this reason, we developed a tapering algorithm for PICU patients.Our hypothesis was that a tapering algorithm would reduce the incidence of IWS in PICU. The aim of the study was to test the algorithm for the tapering of analgesics and sedatives.MATERIALS AND METHODSAlgorithm descriptionMembers of the study group with extensive clinical PICU experience developed the analgosedation tapering algorithm for PICU (MD, FEF, RIH, GAR, and GKB) (Figure 1). The algorithm was designed to support the decision of how fast to taper continuous infusions of opioids and/or benzodiazepines, differentiating between infusions having lasted <5 days and ≥5 days. The study only tested the algorithm on patients receiving continuous infusions for 5 days or more.1FIGUREAlgorithm for tapering of opioids and benzodiazepines.The algorithm uses two validated scales, the Comfort Behavior Scale (Comfort B‐scale)14 and the Withdrawal Assessment Tool‐1 (WAT‐1).11 Before starting to taper infusions, the Comfort B‐Scale should be <17 to ensure that the patients are adequately treated for pain.15 The Comfort B‐Scale has been used in our PICUs for several decades to access pain and distress. It is described as an excellent tool, recommended by a multidisciplinary group of experts from the European Society of Pediatric and Intensive Care (ESPNIC).15,16 The Comfort B‐Scale consists of six behavioral items, alertness, calmness, respiratory response or crying, body movements, facial tension, and muscle tone.17 Each item is rated on a scale from 1 to 5 to describe the intensity of the behavior in question. The sum of the six ratings yields a total score ranging from 6 to 30.17During tapering the Withdrawal Assessment Tool‐1 (WAT‐1) is used to assess for signs or symptoms of withdrawal (WAT‐1 of ≥3), and if indicated directing the nurses to immediately administer a bolus of an opioid (first priority) or a benzodiazepine. WAT‐1 is an 11‐item (12‐point) tool that monitors signs and symptoms of opioid‐ and benzodiazepine‐related IWS in PICU.11,18 A score of ≥3 indicates IWS and the highest possible score is 12. In a recently published systematic review, the WAT‐1 is described as a precise and easy‐to‐use tool to use at the bedside.19The WAT‐1 tool should ensure adequate treatment during the tapering phase while preventing additional medication boluses such as propofol and thiopental. With our algorithm, the tapering phase takes at least 5 days, if already treated with infusions of opioids and benzodiazepines ≥5 days. The algorithm is intended to provide systematic and close patient observation and ensures a longer tapering phase when necessary. A structured form was prepared with the algorithm to calculate medication reduction every 12 h to ease the clinicians’ work and to ensure systematic tapering according to the algorithm (Figure 2).2FIGUREMedication calculation for the tapering algorithm.Study designThis study has a pre‐ and posttest design with the implementation of the algorithm after the pretest data collection was completed. Data from the pretest (baseline phase) are already published.4 Before the inclusion of patients in the pretest, the study group (MD, FEF, RIH, GAR, and GKB) was trained in the use of WAT‐1. We discussed how to use the WAT‐1 score by mail with the originator Dr. Linda Franck. We followed the guidelines for WAT‐1,18 and trained together and individually.After the pretest, the algorithm was introduced to the bedside nurses and the physicians, and they were trained in using the algorithm. After training, the analgosedation tapering algorithm was implemented and tested.The primary outcome was the reduction of IWS, and the secondary outcomes were the length of PICU stay after inclusion, hospital stay, and duration of mechanical ventilation. The trial was registered at CinicalTrials.gov prior to data collection (id. NCT02952846).Study samplesThe proportion of patients with a WAT‐1 score ≥3 was estimated to be around 0.80 in the baseline group. A sample size of 80 children would be adequate to detect a clinically relevant difference of 30% between the baseline group and the intervention group. A two‐sided significance level of 5% was used and the power was set to 80%.Participating children were recruited at two medical‐surgical PICUs with six and three beds respectively at Oslo University Hospital, Norway. The two PICUs are located geographically at different addresses in Oslo. We included two samples, 40 children in the pretest (baseline group), and 40 children in the posttest (intervention group). Children 0–18 years of age requiring mechanical ventilation and analgosedation for ≥5 days were eligible for enrollment. The inclusion criterion was set at ≥5 days as we wished to study patients at increased risk of developing withdrawal. The physician in charge approved patients for inclusion. The exclusion criterion was severe nervous system impairment that could affect the assessment of the sedation level. Patients were included only once. We followed the patients in PICU and the general ward until 3 days after all infusions of analgosedation were discontinued, however, with a limitation of maximum 21 days.Algorithm training programThe bedside nurses that were taking care of the patients in the intervention phase completed a training program after the baseline phase. The program focused on patient assessment using WAT‐1. The nurses trained together and alone and discussed the results of the assessment tool. The program consisted of 72 learning situations where nurses in pairs used the WAT‐1 and scored the patients together. There was 100% agreement in 66 of these scores. In addition, we had 20 learning situations where the bedside nurses practiced by scoring alone and subsequently discussing the results with each other. The algorithm was presented to the nurses and the physicians for comments and feedback prior to the intervention phase.The baseline phaseWe included 40 patients in the baseline group.4 No tool was used by the bedside nurses to assess withdrawal signs and symptoms in the patients at baseline.4 The bedside nurses used their clinical judgment as did the physicians who decided how and when to taper during this phase. Members of the study group (MD, FEF, RIH, GAR, and GKB) WAT‐1 scored the patients twice a day during the baseline.4 The bedside nurses and the physicians were blinded to these scores.The intervention phaseWe included 40 patients in the intervention group where patients were tapered according to the algorithm. The bedside nurses performed WAT‐1 scores as described in the algorithm (Figure 1). The WAT‐1 scores guided the bedside nurses on how to progress with tapering. In the same way, as during baseline, the study group did twice daily WAT‐1 scores generating the numbers going into the results presented in this study. The bedside nurses and the physicians were blinded to these scores as well. The study group was assisted by a few dedicated nurses that helped us to score to complete the study. These assisting nurses were not involved in patient care.EthicsThe Regional Committees for Medical and Health Research Ethics approved this study (2016/135). The parents were carefully informed about the study and of the implementation of the algorithm. The parents provided written informed consent before their child was included in the study.Data collectionDemographic data, clinical characteristics, and medications were recorded from the patient's medical record. In addition, we recorded signs and symptoms that prompted the bedside nurses to administer additional bolus doses of analgesics and/or sedatives. As described in the pre‐test,4 we followed the same procedure using a questionnaire to document the following signs and symptoms: sleep disturbance, tremors, seizures, fever, muscle contraction, hallucinations, crying, agitation/restlessness, sneezing, tachycardia, hypertension, agitation, loose stools, nausea, and vomiting. The nurses also used pro re nata (PRN) orders during this phase and had a choice between opioids, benzodiazepines, and other sedatives such as propofol and thiopental. In contrast to the baseline, the nurses followed the algorithm in the intervention phase that had an opioid as the first priority if they documented a WAT‐1 ≥ 3 (Figure 1).Peak WAT‐1 and SUM WAT‐1We used WAT‐1 as our monitoring tool during the baseline and intervention phase of the study11 to estimate the prevalence of IWS. The tool was translated from English into Norwegian before we started the study.4 We used the newly developed variable in the analysis, the sum WAT‐1 ≥ 3 together with the peak WAT‐1.4 Sum WAT‐1 ≥ 3 is calculated by adding all scores ≥3 during the tapering period of maximum 21 days, and describing the burden of IWS over time. We included two scores per day.4 The Spearman's correlation coefficient between Peak WAT‐1 and Sum WAT‐1 ≥ 3 was 0.8 (CI 0.65–0.89).4 The peak WAT‐1 is used in the analysis with a cutoff value of ≥3 that indicates IWS.11,18 The peak WAT‐1 is the highest score of each patient during the study period in the present study.Data analysisPatient characteristics are reported as medians with interquartile ranges (IQRs) for continuous variables. Frequency and percentage were provided for categorical variables. Differences between the baseline and the intervention group were tested using a Mann–Whitney‐Wilcoxon test, a Fisher's exact test, or a chi‐squared test. The analysis was performed for Intention To Treat (ITT), and Per Protocol (PP) for primary outcomes. Missing values were treated equally in the baseline4 and the intervention groups. We used the closest number in cases with missing values when we summed the WAT‐1 scores ≥3.4The significance level was set at 0.05 for primary outcomes. Analysis was performed using IBM SPSS Statistics version 28 (Citation/ref: IBM Corp. Released 2021. IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp).RESULTSPatient characteristicsAs outlined in Table 1 the groups were similar according to age, diagnosis, and hospital stay. The gender distribution, however, varied as the baseline group had 20/20 girls/boys, and the intervention group had 14/26. The infants in both groups had various congenital malformations such as gastroschisis, omphalocele, diaphragmatic hernia, and esophageal atresia. A few of the patients suffered from leukemia, and mostly the patients in the respiratory failure group suffered from different types of viral pneumonia. We categorized the patients into groups based on the main reason for PICU admission (Table 1). With respect to age, 77.5% of the included patients were less than 2 years old, and 52.5% were younger than 6 months of age. There was one readmission to PICU due to IWS in the baseline group4 and one in the intervention group. The baseline patient scored 8 and the intervention patient scored 7 on return to PICU. Other adverse events were considered not to be related to IWS, because the patients were critically ill and needed additional analgosedation due to complications associated with their illness. Seven patients in the baseline and four patients in the intervention group were reintubated due to respiratory failure. Six baseline patients and four intervention patients had additional surgery, 5 baseline patients and 11 intervention patients required general anesthesia. One baseline patient and three intervention patients needed a change of wound dressing. One baseline patient experienced a pneumothorax. There were no statistical differences between the baseline and the intervention group.1TABLEStudy group characteristics.CharacteristicsBaseline group (N = 40)Intervention group (N = 40)N (%)N (%)p‐valueGenderMale20 (50%)26 (65%)0.178Admission reasonsRespiratory insufficiency19 (47.5%)16 (40%)0.457Postoperative monitoring13 (32.5%)20 (50%)0.112Multiple organ system failure5 (13%)2 (5%)0.432Acute liver failure2 (5%)1 (2.5%)1.000Acute kidney injury1 (2.5%)01.000Circulatory failure01 (2.5%)1.000Median (IQR)Medianp‐value(IQR)Age in months (range)65.50.689(1–21)(0.3–24.8)Weight (kg) (range)6.77.50.946(3.8–11.4)(3.4–12.1)Days on ventilator/NIV (range)990.204(6–13)(6–16.5)Days in PICU1313.50.333(9–19)(10–20.8)Days in hospital32.5260.840(13–51.5)(15.3–44.8)Days of analgosedation treatment18170.780(12.3–30)(11.3–30.5)Days of analgosedation treatment before weaning8.580.842(6–11)(6–10.8)Days of analgosedation weaning13.5130.768(8–21)(6.3–22)Days in PICU after starting weaning570.076(3–10.25)(4.25–10.75)Abbreviations: IQR, interquartile range (25th–75th percentiles); N, number; PICU, pediatric intensive care unit.Comfort B‐scaleThe Comfort B‐Scale score in the morning before tapering started was a median of 12 (IQR 11–13) at baseline and a median of 12.5 (IQR 10.5–15) in the intervention group.Prevalence of IWS before and after the implementation of the algorithmMembers of the study group conducted 1175 WAT‐1 scores at baseline and 1117 WAT‐1 scores during the intervention. The prevalence of IWS, defined as peak WAT‐1 ≥ 3 among these twice daily scores, was significantly lower in the intervention group (52.5%) than at baseline (95%). The median peak WAT‐1 was 5 (IQR 4–6.75) versus 3 (IQR 2–6), with a p‐value = 0.012, (Figure 3). By using sum WAT‐1 ≥ 3, the median was 15.5 (IQR 8.25–39) versus 3 (IQR 0–20), with a p‐value = <0.001, (Figure 4). Eight of the included patients did not receive tapering according to the algorithm. The deviations were incorrect administration of bolus medications, too rapid tapering of opioid and benzodiazepine infusions, and lack of WAT‐1 scores. Our results are primarily based on ITT analysis. However, with PP analysis the median peak WAT‐1 was 2 (IQR 2–5.8) in the intervention group (N = 32), with a p‐value .001, compared to the ITT analysis, and the sum WAT‐1 was median 0 (0–8.8), with p‐value = <.001.3FIGUREPeak WAT‐1 at baseline versus intervention.4FIGURESum WAT‐1 ≥ 3 at baseline versus intervention.Characteristics of analgosedation in the two study groupsAll patients received continuous infusions of opioids and benzodiazepines during their stay in PICU (Table 2). The medication use in the two groups was comparable (Table 2). The use of midazolam after started tapering was slightly less in the intervention group (p = 0.083). The algorithm reduced bolus injections with thiopental and propofol markedly, whereas the bolus injections with opioids increased (Figure 5).2TABLEMedication during the hospital stay.Medication (cumulative dose)Baseline group (N = 40)Intervention group (N = 40)p‐valueMedian (IQR)Median (IQR)Opioid dose before tapering (mg/kg)27.6 (17.5–49.8)24.6 (18.6–38.5)0.810Opioid dose after started tapering (mg/kg)10.6 (4.6–26.2)10.5 (7.7–15.3)0.885Opioid total dose (mg/kg)38.7 (25.4–68.2)35.7 (28.6–57.7)0.715Midazolam dose before tapering (mg/kg)9.8 (3.4–21.7)10.1 (5.1–23.4)0.624Midazolam dose after started tapering (mg/kg)2.2 (0.4–7.3)4.0 (2.0–9.1)0.083Midazolam total dose (mg/kg)13.1 (4.1–29.2)15.3 (7.5–36.4)0.570Dexmedetomidine dose before tapering (πg/kg)0 (0–77.6)4.0 (0–96.0)0.694Dexmedetomidine after started tapering (πg/kg)5.3 (0–71.9)17.6 (0–105.5)0.814Dexmedetomidine total dose (πg/kg)37.3 (0–154.3)67.0 (0–252.9)0.445Thiopental dose before tapering (mg/kg)24.5 (5.8–95.3)14.0 (0–87.9)0.459Thiopental dose after started tapering (mg/kg)6.6 (0–44.4)2.01 (0–16.6)0.348Thiopental total dose (mg/kg)55.2 (8.7–140.6)24.4 (7.2–116.7)0.309Propofol dose before tapering (mg/kg)29.0 (0–93.5)8.7 (0–99.1)0.808Propofol dose after started tapering (mg/kg)7.8 (0–24.6)5.6 (0–21.30)0.729Propofol total dose (mg/kg)43.50 (6.8–125.4)25.6 (3.5–131.9)0.725Clonidine dose before tapering PO (πg/kg)6.4 (0–29.7)5.6 (0–26.1)0.659Clonidine dose after started tapering PO (πg/kg)65.1 (34.1–104.2)57.9 (43.4–120.9)0.969Clonidine total dose PO (πg/kg)86.4 (40.1–120.0)69.8 (46.6–145.2)0.862Abbreviations: IQR, interquartile range (25th–75th percentiles); PO, per oral; opioids are fentanyl converted to morphine equivalents by multiplication by 50 and added together with morphine and ketobemidone (= an opioid equipotent to morphine) in mg.5FIGURETotal numbers of bolus medication during the tapering phase.DISCUSSIONPrevalence of iatrogenic withdrawal syndrome after implementation of a tapering algorithmThe main finding in the present study was a significant reduction in the incidence and severity of IWS in our PICUs after the implementation of the newly developed algorithm for tapering opioids and benzodiazepines. Compared to other studies20,21 the incidences of IWS in our study are high, but then we only included patients with infusions of opioids and benzodiazepines for more than ≥5 days. This group of patients is described in earlier research as a risk group for developing IWS.10 Also, it is always difficult to compare results from studies using different designs, inclusion criteria, samples, scoring tools, and medications.22,23IWS was significantly reduced after the implementation of the algorithm in our study. When using Sum WAT‐1 ≥ 3, illustrating the burden over time, we demonstrated an even more convincing reduction of IWS than using peak WAT‐1. By using the elevated WAT‐1 scores (Sum WAT‐1 ≥ 3) from the daily twice assessments, this enabled us to describe the burden of IWS more convincingly and present a more valid outcome. We believe this is an important finding, as the WAT‐1 tool is sensitive, and there might be a risk of error when using only one WAT‐1 score per patient (Peak WAT‐1). The concerns of using only one WAT‐1 (peak WAT‐1) score per patient have been discussed by others. One study defined IWS by using a WAT‐1 score ≥4 instead of ≥3.24 In other studies, the criteria for IWS was two (not one) WAT‐1 ≥ 3.13,20Of course, in the daily clinical use of WAT‐1, it is only the present score that is important, however, to use SUM WAT‐1 ≥ 3 in our study, has provided the opportunity to observe the burden of IWS over time. We were also able to better utilize the data we collected.In other recently published studies, IWS was also successfully reduced by using weaning protocols.25,26 Elyas et al. found a reduction of IWS on 37% in their pediatric cardiac ICU, and care and sedation quality improved in their postoperative cardiac PICU patients.26 By contrast, we had a 42.5% decrease in IWS in our high‐risk group of patients.The tapering algorithm in the present study enabled early recognition of IWS because by using the WAT‐1 in the algorithm with a cutoff value of ≥3, the bedside intensive care nurses were able to recognize potential IWS immediately in the tapering phase, and quickly manage it with breakthrough bolus medication. The algorithm clearly states that an opioid bolus is the first choice when the WAT‐1 score is ≥3. This can potentially prevent the development of a more long‐lasting burden of IWS. The algorithm reduced bolus injections of thiopental and propofol markedly, while bolus injections with opioids were increased (Figure 5). We infer from these results that the intensive care nurses in the majority of patients adhered to the algorithm.Implementation challengesThe algorithm was designed to facilitate immediate treatment of IWS, so there should ideally be none or very few patients with a Peak WAT‐1 ≥ 3 in the intervention group when the study group scored the patients. It is, however, challenging to obtain this in a real‐world PICU, so 52.5% of the patients had a Peak WAT‐1 ≥ 3 at one point or another. There is no simple explanation as to why the algorithm was not optimally followed in all the included patients in the intervention group. One reason could be that we trained an insufficient number of nurses in using the algorithm, another is that we could have repeated the study protocol and the training program during patient inclusion. We did, but obviously not enough.Our results are based on ITT analysis, but if applying PP analysis,27 the results improve with a reduction in median Peak WAT‐1 score from 3 to 2. So, with better adherence to the algorithm, further improvement may be achieved.Similar challenges of introducing new work routines have been discussed by others. In an observational study from 2019, aiming to assess the implementation of a sedative and analgesic drug rotation protocol in a PICU, the protocol was only followed as intended in 35% of patients.25 Despite holding education sessions on the protocol, the PICU staff did not follow it as intended.25 They concluded that a main obstacle to the successful implementation of a new routine was worry that it would increase the workload for physicians and nurses.25 We were also concerned about this, so accordingly we developed the form with the medical calculation (Figure 2). The intention was to simplify the work for the clinicians when handing over the patients between shifts during the tapering phase. We think it is important to make new routines as simple as possible so that the PICU staff easily can use them. With continued use of the algorithm and staff training, we believe we will improve adherence to the algorithm.Secondary outcomesOur secondary outcomes were the duration of PICU stay, hospital stay, and duration of mechanical ventilation. By comparing the baseline and the intervention group in the present study, we did not find significant differences in these variables. Our concern that the algorithm using at least five days to taper the analgosedation after prolonged administration would cause prolonged PICU LOS was not supported by our numbers. A multicenter study with a 64.4% incidence of IWS, found a significant difference between the patients who did and did not develop IWS regarding several variables such as PICU stay, hospital stay, and days of ventilator treatment.12 The patients that developed IWS had longer hospital stay and PICU stay and needed more days of ventilator treatment.12 Similar findings are also published by others, with longer duration of analgesics, longer duration of mechanical ventilation, and longer PICU stay.20,28,29 So, we assume that it is the presence of IWS that increases LOS.LIMITATIONSOur small‐scale study has some limitations that must be mentioned. First, the design is a pre‐ and post‐design with a baseline group and an intervention group, not a randomized controlled trial. The nature of the intervention makes randomization very difficult, maybe impossible. Our results have to be interpreted with caution in light of the fact that there might have been other causes that have led to our improving results. Another limitation is the lack of adherence to the algorithm. Further, other conditions such as delirium can be confused with withdrawal. The occurrence of delirium was not examined in the study. Finally, the samples are small, and we tried but did not succeed in bringing in more PICUs in other hospitals in our study.CONCLUSIONSImplementation of an algorithm for tapering analgosedation significantly reduced IWS in our sample. The length of hospital stay, PICU stay after inclusion, and duration of mechanical ventilation did not differ significantly between the baseline group and the intervention group. By including WAT‐1 in the algorithm and letting the WAT‐1 control the tapering rate, we succeeded in reducing the incidence of IWS. It seems very important to use a systematic approach and a validated tool in the treatment of PICU patients when they are in a tapering phase of opioids and benzodiazepines.AUTHOR CONTRIBUTIONSAll the co‐authors have contributed throughout the process of writing this article.ACKNOWLEDGMENTSThe authors would like to acknowledge the nurses and physicians involved in the patients’ ward at Rikshospitalet and Ullevaal, Oslo University Hospital, for collaboration in the conduct of this study. Thanks to Dr. Linda Frank for the permission to use WAT‐1 in the study.CONFLICT OF INTEREST STATEMENTThe authors have no conflicts of interest.DATA AVAILABILITY STATEMENTThe data that support the findings of this study are available from the corresponding author upon reasonable request.REFERENCESAnand KJ, Willson DF, Berger J, et al. National Institute of child H, human development collaborative pediatric critical care research N. tolerance and withdrawal from prolonged opioid use in critically ill children. Pediatrics. 2010;125:e1208‐e1225.Birchley G. Opioid and benzodiazepine withdrawal syndromes in the paediatric intensive care unit: a review of recent literature. Nurs Crit Care. 2009;14:26‐37.Vet NJ, Ista E, de Wildt SN, van Dijk M, Tibboel D, de Hoog M. Optimal sedation in pediatric intensive care patients: a systematic review. Intensive Care Med. 2013;39:1524‐1534.Dokken M, Rustoen T, Diep LM, et al. Iatrogenic withdrawal syndrome frequently occurs in paediatric intensive care without algorithm for tapering of analgosedation. Acta Anaesthesiol Scand. 2021;65:928‐935.LaRosa JM, Aponte‐Patel L. Iatrogenic withdrawal syndrome: a review of pathophysiology, prevention, and treatment. Curr Pediatr Rep. 2019;7:12‐19.Biswas AK, Feldman BL, Davis DH, Zintz EA. Myocardial ischemia as a result of severe benzodiazepine and opioid withdrawal. Clin Toxicol (Phila). 2005;43:207‐209.Mencia S, Sanavia E, Fernandez S, et al. Evaluation of sedative and analgesic drug rotation protocol to decrease withdrawal syndrome in critically ill children with prolonged sedation. Pediatr Crit Care Med. 2018;19(6 Supplement 1):200.Jenkins IA, Playfor SD, Bevan C, Davies G, Wolf AR. Current United Kingdom sedation practice in pediatric intensive care. Paediatr Anaesth. 2007;17:675‐683.Whelan KT, Heckmann MK, Lincoln PA, Hamilton SM. Pediatric withdrawal identification and management. J Pediatr Intensive Care. 2015;4:73‐78.Best KM, Boullata JI, Curley MA. Risk factors associated with iatrogenic opioid and benzodiazepine withdrawal in critically ill pediatric patients: a systematic review and conceptual model. Pediatr Crit Care Med. 2015;16:175‐183.Franck LS, Scoppettuolo LA, Wypij D, Curley MA. Validity and generalizability of the withdrawal assessment tool‐1 (WAT‐1) for monitoring iatrogenic withdrawal syndrome in pediatric patients. Pain. 2012;153:142‐148.Amigoni A, Mondardini MC, Vittadello I, et al. Withdrawal assessment tool‐1 monitoring in PICU: a multicenter study on iatrogenic withdrawal syndrome. Pediatr Crit Care Med. 2017;18:e86‐e91.Best KM, Wypij D, Asaro LA, Curley MA, Randomized Evaluation of Sedation Titration For Respiratory Failure Study I. Patient, process, and system predictors of iatrogenic withdrawal syndrome in critically ill children. Crit Care Med. 2017;45:e7‐e15.Ista E, van Dijk M, Tibboel D, de Hoog M. Assessment of sedation levels in pediatric intensive care patients can be improved by using the COMFORT "behavior" scale. Pediatr Crit Care Med. 2005;6:58‐63.Dorfman TL, Sumamo Schellenberg E, Rempel GR, Scott SD, Hartling L. An evaluation of instruments for scoring physiological and behavioral cues of pain, non‐pain related distress, and adequacy of analgesia and sedation in pediatric mechanically ventilated patients: a systematic review. Int J Nurs Stud. 2014;51:654‐676.Harris J, Ramelet AS, van Dijk M, et al. Clinical recommendations for pain, sedation, withdrawal and delirium assessment in critically ill infants and children: an ESPNIC position statement for healthcare professionals. Intensive Care Med. 2016;42:972‐986.Boerlage AA, Ista E, Duivenvoorden HJ, de Wildt SN, Tibboel D, van Dijk M. The COMFORT behaviour scale detects clinically meaningful effects of analgesic and sedative treatment. Eur J Pain. 2015;19:473‐479.Franck LS, Harris SK, Soetenga DJ, Amling JK, Curley MA. The Withdrawal Assessment Tool‐1 (WAT‐1): an assessment instrument for monitoring opioid and benzodiazepine withdrawal symptoms in pediatric patients. Pediatr Crit Care Med. 2008;9:573‐580.Zaccagnini M, Ataman R, Nonoyama ML. The withdrawal assessment tool to identify iatrogenic withdrawal symptoms in critically ill paediatric patients: a COSMIN systematic review of measurement properties. J Eval Clin Pract. 2021;27:976‐988.Habib E, Almakadma AH, Albarazi M, et al. Iatrogenic withdrawal syndrome in the pediatric cardiac intensive care unit: incidence, risk factors and outcome. J Saudi Heart Assoc. 2021;33:251‐260.Amigoni A, Vettore E, Brugnolaro V, et al. High doses of benzodiazepine predict analgesic and sedative drug withdrawal syndrome in paediatric intensive care patients. Acta Paediatr. 2014;103:e538‐e543.Ávila‐Alzate JA, Gómez‐Salgado J, Romero‐Martín M, Martínez‐Isasi S, Navarro‐Abal Y, Fernández‐García D. Assessment and treatment of the withdrawal syndrome in paediatric intensive care units: systematic review. Medicine (Baltimore). 2020;99:e18502.Bichaff P, Setani KT, Motta EHG, Delgado AF, Carvalho WB, Luglio M. Opioid tapering and weaning protocols in pediatric critical care units: a systematic review. Rev Assoc Med Bras. 1992;2018(64):909‐915.Amirnovin R, Sanchez‐Pinto LN, Okuhara C, et al. Implementation of a risk‐stratified opioid and benzodiazepine weaning protocol in a pediatric cardiac ICU. Pediatr Crit Care Med. 2018;19:1024‐1032.Sanavia E, Mencía S, Lafever SN, Solana MJ, Garcia M, López‐Herce J. Sedative and analgesic drug rotation protocol in critically ill children with prolonged sedation: evaluation of implementation and efficacy to reduce withdrawal syndrome. Pediatr Crit Care Med. 2019;20:1111‐1117.Elyas M, Haggag S, Atef A. 1306 Implementation of Withdrawal Assessment Tool and Weaning Protocol to Reduce Iatrogenic Withdrawal Syndrome in Pediatric Cardiac ICU: a Quality Improvement Project. BMJ Publishing Group Ltd; 2022.Hernán MA, Hernández‐Díaz S. Beyond the intention‐to‐treat in comparative effectiveness research. Clin Trials. 2012;9:48‐55.da Silva PS, Reis ME, Fonseca TS, Fonseca MC. Opioid and benzodiazepine withdrawal syndrome in PICU patients: which risk factors matter? J Addict Med. 2016;10:110‐116.Geslain G, Ponsin P, Lãzãrescu AM, et al. Incidence of iatrogenic withdrawal syndrome and associated factors in surgical pediatric intensive care. Arch Pediatr. 2022;30:14‐19. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Anaesthesiologica Scandinavica Wiley

Implementation of an algorithm for tapering analgosedation reduces iatrogenic withdrawal syndrome in pediatric intensive care

Download PDF
10 pages

Loading next page...
 
/lp/wiley/implementation-of-an-algorithm-for-tapering-analgosedation-reduces-0IWC56JAFy

References (30)

Publisher
Wiley
Copyright
© 2023 The Acta Anaesthesiologica Scandinavica Foundation
ISSN
0001-5172
eISSN
1399-6576
DOI
10.1111/aas.14288
Publisher site
See Article on Publisher Site

Abstract

Editorial CommentThe study proposes a simple algorithm to taper analgosedation in a pediatric intensive care setting to reduce iatrogenic withdrawal syndrome (IWS) in children requiring sedation for more than 5 days. The use of peak and sum WAT scores is a useful indicator of the severity and prevalence of IWS. The algorithm still requires validation in different pediatric intensive care settings.INTRODUCTIONProper analgosedation is essential in the treatment of critically ill mechanically ventilated patients in pediatric intensive care units (PICUs). Analgosedation where pain is addressed before sedation is essential for safe and respectful care.1 The most common agents used for the past decades are opioids such as fentanyl and morphine, and sedatives such as midazolam.2 Pediatric intensive care patients need continuous infusions of these medications to maintain an adequate level of comfort. Further, intensive care nurses must prevent unplanned extubation and loss of vascular accesses. However, the use of these medications over time may lead to side effects such as iatrogenic withdrawal syndrome (IWS), prolonged mechanical ventilation (MV), and increased PICU length of stay (LOS).3 If analgesics or sedatives are tapered too rapidly, the patient is at risk of IWS, which is determined by a typical pattern of signs and symptoms showing central nervous system drug effects.1 The symptoms cause discomfort during the tapering phase. In an earlier study, we found that tachycardia, agitation, sleep disturbance, crying, and tremors were the most frequent signs and symptoms in the current study units and that they prompted additional bolus medication.4 IWS can result in continued and adverse health consequences for the patients, including seizures, myocardial ischemia, and increased length of PICU stay and hospital stay.5,6Different strategies have been described to reduce the side effects of IWS, such as following sedation protocols7 and performing daily sedation interruption.8 Nonetheless, assessment, prevention, and management of IWS continue to be challenging.9 Best et al. identified risk factors associated with IWS and found that duration of therapy and cumulative doses of opioids and benzodiazepines are the most predictive factor of IWS.10 They also highlighted the need to further explore process, and system variables, such as sedation/IWS assessment, and protocol adherence.10 The incidence of IWS in recovery is highly variable in different studies, ranging between 47% and 95%,4,11–13 and for this reason, we developed a tapering algorithm for PICU patients.Our hypothesis was that a tapering algorithm would reduce the incidence of IWS in PICU. The aim of the study was to test the algorithm for the tapering of analgesics and sedatives.MATERIALS AND METHODSAlgorithm descriptionMembers of the study group with extensive clinical PICU experience developed the analgosedation tapering algorithm for PICU (MD, FEF, RIH, GAR, and GKB) (Figure 1). The algorithm was designed to support the decision of how fast to taper continuous infusions of opioids and/or benzodiazepines, differentiating between infusions having lasted <5 days and ≥5 days. The study only tested the algorithm on patients receiving continuous infusions for 5 days or more.1FIGUREAlgorithm for tapering of opioids and benzodiazepines.The algorithm uses two validated scales, the Comfort Behavior Scale (Comfort B‐scale)14 and the Withdrawal Assessment Tool‐1 (WAT‐1).11 Before starting to taper infusions, the Comfort B‐Scale should be <17 to ensure that the patients are adequately treated for pain.15 The Comfort B‐Scale has been used in our PICUs for several decades to access pain and distress. It is described as an excellent tool, recommended by a multidisciplinary group of experts from the European Society of Pediatric and Intensive Care (ESPNIC).15,16 The Comfort B‐Scale consists of six behavioral items, alertness, calmness, respiratory response or crying, body movements, facial tension, and muscle tone.17 Each item is rated on a scale from 1 to 5 to describe the intensity of the behavior in question. The sum of the six ratings yields a total score ranging from 6 to 30.17During tapering the Withdrawal Assessment Tool‐1 (WAT‐1) is used to assess for signs or symptoms of withdrawal (WAT‐1 of ≥3), and if indicated directing the nurses to immediately administer a bolus of an opioid (first priority) or a benzodiazepine. WAT‐1 is an 11‐item (12‐point) tool that monitors signs and symptoms of opioid‐ and benzodiazepine‐related IWS in PICU.11,18 A score of ≥3 indicates IWS and the highest possible score is 12. In a recently published systematic review, the WAT‐1 is described as a precise and easy‐to‐use tool to use at the bedside.19The WAT‐1 tool should ensure adequate treatment during the tapering phase while preventing additional medication boluses such as propofol and thiopental. With our algorithm, the tapering phase takes at least 5 days, if already treated with infusions of opioids and benzodiazepines ≥5 days. The algorithm is intended to provide systematic and close patient observation and ensures a longer tapering phase when necessary. A structured form was prepared with the algorithm to calculate medication reduction every 12 h to ease the clinicians’ work and to ensure systematic tapering according to the algorithm (Figure 2).2FIGUREMedication calculation for the tapering algorithm.Study designThis study has a pre‐ and posttest design with the implementation of the algorithm after the pretest data collection was completed. Data from the pretest (baseline phase) are already published.4 Before the inclusion of patients in the pretest, the study group (MD, FEF, RIH, GAR, and GKB) was trained in the use of WAT‐1. We discussed how to use the WAT‐1 score by mail with the originator Dr. Linda Franck. We followed the guidelines for WAT‐1,18 and trained together and individually.After the pretest, the algorithm was introduced to the bedside nurses and the physicians, and they were trained in using the algorithm. After training, the analgosedation tapering algorithm was implemented and tested.The primary outcome was the reduction of IWS, and the secondary outcomes were the length of PICU stay after inclusion, hospital stay, and duration of mechanical ventilation. The trial was registered at CinicalTrials.gov prior to data collection (id. NCT02952846).Study samplesThe proportion of patients with a WAT‐1 score ≥3 was estimated to be around 0.80 in the baseline group. A sample size of 80 children would be adequate to detect a clinically relevant difference of 30% between the baseline group and the intervention group. A two‐sided significance level of 5% was used and the power was set to 80%.Participating children were recruited at two medical‐surgical PICUs with six and three beds respectively at Oslo University Hospital, Norway. The two PICUs are located geographically at different addresses in Oslo. We included two samples, 40 children in the pretest (baseline group), and 40 children in the posttest (intervention group). Children 0–18 years of age requiring mechanical ventilation and analgosedation for ≥5 days were eligible for enrollment. The inclusion criterion was set at ≥5 days as we wished to study patients at increased risk of developing withdrawal. The physician in charge approved patients for inclusion. The exclusion criterion was severe nervous system impairment that could affect the assessment of the sedation level. Patients were included only once. We followed the patients in PICU and the general ward until 3 days after all infusions of analgosedation were discontinued, however, with a limitation of maximum 21 days.Algorithm training programThe bedside nurses that were taking care of the patients in the intervention phase completed a training program after the baseline phase. The program focused on patient assessment using WAT‐1. The nurses trained together and alone and discussed the results of the assessment tool. The program consisted of 72 learning situations where nurses in pairs used the WAT‐1 and scored the patients together. There was 100% agreement in 66 of these scores. In addition, we had 20 learning situations where the bedside nurses practiced by scoring alone and subsequently discussing the results with each other. The algorithm was presented to the nurses and the physicians for comments and feedback prior to the intervention phase.The baseline phaseWe included 40 patients in the baseline group.4 No tool was used by the bedside nurses to assess withdrawal signs and symptoms in the patients at baseline.4 The bedside nurses used their clinical judgment as did the physicians who decided how and when to taper during this phase. Members of the study group (MD, FEF, RIH, GAR, and GKB) WAT‐1 scored the patients twice a day during the baseline.4 The bedside nurses and the physicians were blinded to these scores.The intervention phaseWe included 40 patients in the intervention group where patients were tapered according to the algorithm. The bedside nurses performed WAT‐1 scores as described in the algorithm (Figure 1). The WAT‐1 scores guided the bedside nurses on how to progress with tapering. In the same way, as during baseline, the study group did twice daily WAT‐1 scores generating the numbers going into the results presented in this study. The bedside nurses and the physicians were blinded to these scores as well. The study group was assisted by a few dedicated nurses that helped us to score to complete the study. These assisting nurses were not involved in patient care.EthicsThe Regional Committees for Medical and Health Research Ethics approved this study (2016/135). The parents were carefully informed about the study and of the implementation of the algorithm. The parents provided written informed consent before their child was included in the study.Data collectionDemographic data, clinical characteristics, and medications were recorded from the patient's medical record. In addition, we recorded signs and symptoms that prompted the bedside nurses to administer additional bolus doses of analgesics and/or sedatives. As described in the pre‐test,4 we followed the same procedure using a questionnaire to document the following signs and symptoms: sleep disturbance, tremors, seizures, fever, muscle contraction, hallucinations, crying, agitation/restlessness, sneezing, tachycardia, hypertension, agitation, loose stools, nausea, and vomiting. The nurses also used pro re nata (PRN) orders during this phase and had a choice between opioids, benzodiazepines, and other sedatives such as propofol and thiopental. In contrast to the baseline, the nurses followed the algorithm in the intervention phase that had an opioid as the first priority if they documented a WAT‐1 ≥ 3 (Figure 1).Peak WAT‐1 and SUM WAT‐1We used WAT‐1 as our monitoring tool during the baseline and intervention phase of the study11 to estimate the prevalence of IWS. The tool was translated from English into Norwegian before we started the study.4 We used the newly developed variable in the analysis, the sum WAT‐1 ≥ 3 together with the peak WAT‐1.4 Sum WAT‐1 ≥ 3 is calculated by adding all scores ≥3 during the tapering period of maximum 21 days, and describing the burden of IWS over time. We included two scores per day.4 The Spearman's correlation coefficient between Peak WAT‐1 and Sum WAT‐1 ≥ 3 was 0.8 (CI 0.65–0.89).4 The peak WAT‐1 is used in the analysis with a cutoff value of ≥3 that indicates IWS.11,18 The peak WAT‐1 is the highest score of each patient during the study period in the present study.Data analysisPatient characteristics are reported as medians with interquartile ranges (IQRs) for continuous variables. Frequency and percentage were provided for categorical variables. Differences between the baseline and the intervention group were tested using a Mann–Whitney‐Wilcoxon test, a Fisher's exact test, or a chi‐squared test. The analysis was performed for Intention To Treat (ITT), and Per Protocol (PP) for primary outcomes. Missing values were treated equally in the baseline4 and the intervention groups. We used the closest number in cases with missing values when we summed the WAT‐1 scores ≥3.4The significance level was set at 0.05 for primary outcomes. Analysis was performed using IBM SPSS Statistics version 28 (Citation/ref: IBM Corp. Released 2021. IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp).RESULTSPatient characteristicsAs outlined in Table 1 the groups were similar according to age, diagnosis, and hospital stay. The gender distribution, however, varied as the baseline group had 20/20 girls/boys, and the intervention group had 14/26. The infants in both groups had various congenital malformations such as gastroschisis, omphalocele, diaphragmatic hernia, and esophageal atresia. A few of the patients suffered from leukemia, and mostly the patients in the respiratory failure group suffered from different types of viral pneumonia. We categorized the patients into groups based on the main reason for PICU admission (Table 1). With respect to age, 77.5% of the included patients were less than 2 years old, and 52.5% were younger than 6 months of age. There was one readmission to PICU due to IWS in the baseline group4 and one in the intervention group. The baseline patient scored 8 and the intervention patient scored 7 on return to PICU. Other adverse events were considered not to be related to IWS, because the patients were critically ill and needed additional analgosedation due to complications associated with their illness. Seven patients in the baseline and four patients in the intervention group were reintubated due to respiratory failure. Six baseline patients and four intervention patients had additional surgery, 5 baseline patients and 11 intervention patients required general anesthesia. One baseline patient and three intervention patients needed a change of wound dressing. One baseline patient experienced a pneumothorax. There were no statistical differences between the baseline and the intervention group.1TABLEStudy group characteristics.CharacteristicsBaseline group (N = 40)Intervention group (N = 40)N (%)N (%)p‐valueGenderMale20 (50%)26 (65%)0.178Admission reasonsRespiratory insufficiency19 (47.5%)16 (40%)0.457Postoperative monitoring13 (32.5%)20 (50%)0.112Multiple organ system failure5 (13%)2 (5%)0.432Acute liver failure2 (5%)1 (2.5%)1.000Acute kidney injury1 (2.5%)01.000Circulatory failure01 (2.5%)1.000Median (IQR)Medianp‐value(IQR)Age in months (range)65.50.689(1–21)(0.3–24.8)Weight (kg) (range)6.77.50.946(3.8–11.4)(3.4–12.1)Days on ventilator/NIV (range)990.204(6–13)(6–16.5)Days in PICU1313.50.333(9–19)(10–20.8)Days in hospital32.5260.840(13–51.5)(15.3–44.8)Days of analgosedation treatment18170.780(12.3–30)(11.3–30.5)Days of analgosedation treatment before weaning8.580.842(6–11)(6–10.8)Days of analgosedation weaning13.5130.768(8–21)(6.3–22)Days in PICU after starting weaning570.076(3–10.25)(4.25–10.75)Abbreviations: IQR, interquartile range (25th–75th percentiles); N, number; PICU, pediatric intensive care unit.Comfort B‐scaleThe Comfort B‐Scale score in the morning before tapering started was a median of 12 (IQR 11–13) at baseline and a median of 12.5 (IQR 10.5–15) in the intervention group.Prevalence of IWS before and after the implementation of the algorithmMembers of the study group conducted 1175 WAT‐1 scores at baseline and 1117 WAT‐1 scores during the intervention. The prevalence of IWS, defined as peak WAT‐1 ≥ 3 among these twice daily scores, was significantly lower in the intervention group (52.5%) than at baseline (95%). The median peak WAT‐1 was 5 (IQR 4–6.75) versus 3 (IQR 2–6), with a p‐value = 0.012, (Figure 3). By using sum WAT‐1 ≥ 3, the median was 15.5 (IQR 8.25–39) versus 3 (IQR 0–20), with a p‐value = <0.001, (Figure 4). Eight of the included patients did not receive tapering according to the algorithm. The deviations were incorrect administration of bolus medications, too rapid tapering of opioid and benzodiazepine infusions, and lack of WAT‐1 scores. Our results are primarily based on ITT analysis. However, with PP analysis the median peak WAT‐1 was 2 (IQR 2–5.8) in the intervention group (N = 32), with a p‐value .001, compared to the ITT analysis, and the sum WAT‐1 was median 0 (0–8.8), with p‐value = <.001.3FIGUREPeak WAT‐1 at baseline versus intervention.4FIGURESum WAT‐1 ≥ 3 at baseline versus intervention.Characteristics of analgosedation in the two study groupsAll patients received continuous infusions of opioids and benzodiazepines during their stay in PICU (Table 2). The medication use in the two groups was comparable (Table 2). The use of midazolam after started tapering was slightly less in the intervention group (p = 0.083). The algorithm reduced bolus injections with thiopental and propofol markedly, whereas the bolus injections with opioids increased (Figure 5).2TABLEMedication during the hospital stay.Medication (cumulative dose)Baseline group (N = 40)Intervention group (N = 40)p‐valueMedian (IQR)Median (IQR)Opioid dose before tapering (mg/kg)27.6 (17.5–49.8)24.6 (18.6–38.5)0.810Opioid dose after started tapering (mg/kg)10.6 (4.6–26.2)10.5 (7.7–15.3)0.885Opioid total dose (mg/kg)38.7 (25.4–68.2)35.7 (28.6–57.7)0.715Midazolam dose before tapering (mg/kg)9.8 (3.4–21.7)10.1 (5.1–23.4)0.624Midazolam dose after started tapering (mg/kg)2.2 (0.4–7.3)4.0 (2.0–9.1)0.083Midazolam total dose (mg/kg)13.1 (4.1–29.2)15.3 (7.5–36.4)0.570Dexmedetomidine dose before tapering (πg/kg)0 (0–77.6)4.0 (0–96.0)0.694Dexmedetomidine after started tapering (πg/kg)5.3 (0–71.9)17.6 (0–105.5)0.814Dexmedetomidine total dose (πg/kg)37.3 (0–154.3)67.0 (0–252.9)0.445Thiopental dose before tapering (mg/kg)24.5 (5.8–95.3)14.0 (0–87.9)0.459Thiopental dose after started tapering (mg/kg)6.6 (0–44.4)2.01 (0–16.6)0.348Thiopental total dose (mg/kg)55.2 (8.7–140.6)24.4 (7.2–116.7)0.309Propofol dose before tapering (mg/kg)29.0 (0–93.5)8.7 (0–99.1)0.808Propofol dose after started tapering (mg/kg)7.8 (0–24.6)5.6 (0–21.30)0.729Propofol total dose (mg/kg)43.50 (6.8–125.4)25.6 (3.5–131.9)0.725Clonidine dose before tapering PO (πg/kg)6.4 (0–29.7)5.6 (0–26.1)0.659Clonidine dose after started tapering PO (πg/kg)65.1 (34.1–104.2)57.9 (43.4–120.9)0.969Clonidine total dose PO (πg/kg)86.4 (40.1–120.0)69.8 (46.6–145.2)0.862Abbreviations: IQR, interquartile range (25th–75th percentiles); PO, per oral; opioids are fentanyl converted to morphine equivalents by multiplication by 50 and added together with morphine and ketobemidone (= an opioid equipotent to morphine) in mg.5FIGURETotal numbers of bolus medication during the tapering phase.DISCUSSIONPrevalence of iatrogenic withdrawal syndrome after implementation of a tapering algorithmThe main finding in the present study was a significant reduction in the incidence and severity of IWS in our PICUs after the implementation of the newly developed algorithm for tapering opioids and benzodiazepines. Compared to other studies20,21 the incidences of IWS in our study are high, but then we only included patients with infusions of opioids and benzodiazepines for more than ≥5 days. This group of patients is described in earlier research as a risk group for developing IWS.10 Also, it is always difficult to compare results from studies using different designs, inclusion criteria, samples, scoring tools, and medications.22,23IWS was significantly reduced after the implementation of the algorithm in our study. When using Sum WAT‐1 ≥ 3, illustrating the burden over time, we demonstrated an even more convincing reduction of IWS than using peak WAT‐1. By using the elevated WAT‐1 scores (Sum WAT‐1 ≥ 3) from the daily twice assessments, this enabled us to describe the burden of IWS more convincingly and present a more valid outcome. We believe this is an important finding, as the WAT‐1 tool is sensitive, and there might be a risk of error when using only one WAT‐1 score per patient (Peak WAT‐1). The concerns of using only one WAT‐1 (peak WAT‐1) score per patient have been discussed by others. One study defined IWS by using a WAT‐1 score ≥4 instead of ≥3.24 In other studies, the criteria for IWS was two (not one) WAT‐1 ≥ 3.13,20Of course, in the daily clinical use of WAT‐1, it is only the present score that is important, however, to use SUM WAT‐1 ≥ 3 in our study, has provided the opportunity to observe the burden of IWS over time. We were also able to better utilize the data we collected.In other recently published studies, IWS was also successfully reduced by using weaning protocols.25,26 Elyas et al. found a reduction of IWS on 37% in their pediatric cardiac ICU, and care and sedation quality improved in their postoperative cardiac PICU patients.26 By contrast, we had a 42.5% decrease in IWS in our high‐risk group of patients.The tapering algorithm in the present study enabled early recognition of IWS because by using the WAT‐1 in the algorithm with a cutoff value of ≥3, the bedside intensive care nurses were able to recognize potential IWS immediately in the tapering phase, and quickly manage it with breakthrough bolus medication. The algorithm clearly states that an opioid bolus is the first choice when the WAT‐1 score is ≥3. This can potentially prevent the development of a more long‐lasting burden of IWS. The algorithm reduced bolus injections of thiopental and propofol markedly, while bolus injections with opioids were increased (Figure 5). We infer from these results that the intensive care nurses in the majority of patients adhered to the algorithm.Implementation challengesThe algorithm was designed to facilitate immediate treatment of IWS, so there should ideally be none or very few patients with a Peak WAT‐1 ≥ 3 in the intervention group when the study group scored the patients. It is, however, challenging to obtain this in a real‐world PICU, so 52.5% of the patients had a Peak WAT‐1 ≥ 3 at one point or another. There is no simple explanation as to why the algorithm was not optimally followed in all the included patients in the intervention group. One reason could be that we trained an insufficient number of nurses in using the algorithm, another is that we could have repeated the study protocol and the training program during patient inclusion. We did, but obviously not enough.Our results are based on ITT analysis, but if applying PP analysis,27 the results improve with a reduction in median Peak WAT‐1 score from 3 to 2. So, with better adherence to the algorithm, further improvement may be achieved.Similar challenges of introducing new work routines have been discussed by others. In an observational study from 2019, aiming to assess the implementation of a sedative and analgesic drug rotation protocol in a PICU, the protocol was only followed as intended in 35% of patients.25 Despite holding education sessions on the protocol, the PICU staff did not follow it as intended.25 They concluded that a main obstacle to the successful implementation of a new routine was worry that it would increase the workload for physicians and nurses.25 We were also concerned about this, so accordingly we developed the form with the medical calculation (Figure 2). The intention was to simplify the work for the clinicians when handing over the patients between shifts during the tapering phase. We think it is important to make new routines as simple as possible so that the PICU staff easily can use them. With continued use of the algorithm and staff training, we believe we will improve adherence to the algorithm.Secondary outcomesOur secondary outcomes were the duration of PICU stay, hospital stay, and duration of mechanical ventilation. By comparing the baseline and the intervention group in the present study, we did not find significant differences in these variables. Our concern that the algorithm using at least five days to taper the analgosedation after prolonged administration would cause prolonged PICU LOS was not supported by our numbers. A multicenter study with a 64.4% incidence of IWS, found a significant difference between the patients who did and did not develop IWS regarding several variables such as PICU stay, hospital stay, and days of ventilator treatment.12 The patients that developed IWS had longer hospital stay and PICU stay and needed more days of ventilator treatment.12 Similar findings are also published by others, with longer duration of analgesics, longer duration of mechanical ventilation, and longer PICU stay.20,28,29 So, we assume that it is the presence of IWS that increases LOS.LIMITATIONSOur small‐scale study has some limitations that must be mentioned. First, the design is a pre‐ and post‐design with a baseline group and an intervention group, not a randomized controlled trial. The nature of the intervention makes randomization very difficult, maybe impossible. Our results have to be interpreted with caution in light of the fact that there might have been other causes that have led to our improving results. Another limitation is the lack of adherence to the algorithm. Further, other conditions such as delirium can be confused with withdrawal. The occurrence of delirium was not examined in the study. Finally, the samples are small, and we tried but did not succeed in bringing in more PICUs in other hospitals in our study.CONCLUSIONSImplementation of an algorithm for tapering analgosedation significantly reduced IWS in our sample. The length of hospital stay, PICU stay after inclusion, and duration of mechanical ventilation did not differ significantly between the baseline group and the intervention group. By including WAT‐1 in the algorithm and letting the WAT‐1 control the tapering rate, we succeeded in reducing the incidence of IWS. It seems very important to use a systematic approach and a validated tool in the treatment of PICU patients when they are in a tapering phase of opioids and benzodiazepines.AUTHOR CONTRIBUTIONSAll the co‐authors have contributed throughout the process of writing this article.ACKNOWLEDGMENTSThe authors would like to acknowledge the nurses and physicians involved in the patients’ ward at Rikshospitalet and Ullevaal, Oslo University Hospital, for collaboration in the conduct of this study. Thanks to Dr. Linda Frank for the permission to use WAT‐1 in the study.CONFLICT OF INTEREST STATEMENTThe authors have no conflicts of interest.DATA AVAILABILITY STATEMENTThe data that support the findings of this study are available from the corresponding author upon reasonable request.REFERENCESAnand KJ, Willson DF, Berger J, et al. National Institute of child H, human development collaborative pediatric critical care research N. tolerance and withdrawal from prolonged opioid use in critically ill children. Pediatrics. 2010;125:e1208‐e1225.Birchley G. Opioid and benzodiazepine withdrawal syndromes in the paediatric intensive care unit: a review of recent literature. Nurs Crit Care. 2009;14:26‐37.Vet NJ, Ista E, de Wildt SN, van Dijk M, Tibboel D, de Hoog M. Optimal sedation in pediatric intensive care patients: a systematic review. Intensive Care Med. 2013;39:1524‐1534.Dokken M, Rustoen T, Diep LM, et al. Iatrogenic withdrawal syndrome frequently occurs in paediatric intensive care without algorithm for tapering of analgosedation. Acta Anaesthesiol Scand. 2021;65:928‐935.LaRosa JM, Aponte‐Patel L. Iatrogenic withdrawal syndrome: a review of pathophysiology, prevention, and treatment. Curr Pediatr Rep. 2019;7:12‐19.Biswas AK, Feldman BL, Davis DH, Zintz EA. Myocardial ischemia as a result of severe benzodiazepine and opioid withdrawal. Clin Toxicol (Phila). 2005;43:207‐209.Mencia S, Sanavia E, Fernandez S, et al. Evaluation of sedative and analgesic drug rotation protocol to decrease withdrawal syndrome in critically ill children with prolonged sedation. Pediatr Crit Care Med. 2018;19(6 Supplement 1):200.Jenkins IA, Playfor SD, Bevan C, Davies G, Wolf AR. Current United Kingdom sedation practice in pediatric intensive care. Paediatr Anaesth. 2007;17:675‐683.Whelan KT, Heckmann MK, Lincoln PA, Hamilton SM. Pediatric withdrawal identification and management. J Pediatr Intensive Care. 2015;4:73‐78.Best KM, Boullata JI, Curley MA. Risk factors associated with iatrogenic opioid and benzodiazepine withdrawal in critically ill pediatric patients: a systematic review and conceptual model. Pediatr Crit Care Med. 2015;16:175‐183.Franck LS, Scoppettuolo LA, Wypij D, Curley MA. Validity and generalizability of the withdrawal assessment tool‐1 (WAT‐1) for monitoring iatrogenic withdrawal syndrome in pediatric patients. Pain. 2012;153:142‐148.Amigoni A, Mondardini MC, Vittadello I, et al. Withdrawal assessment tool‐1 monitoring in PICU: a multicenter study on iatrogenic withdrawal syndrome. Pediatr Crit Care Med. 2017;18:e86‐e91.Best KM, Wypij D, Asaro LA, Curley MA, Randomized Evaluation of Sedation Titration For Respiratory Failure Study I. Patient, process, and system predictors of iatrogenic withdrawal syndrome in critically ill children. Crit Care Med. 2017;45:e7‐e15.Ista E, van Dijk M, Tibboel D, de Hoog M. Assessment of sedation levels in pediatric intensive care patients can be improved by using the COMFORT "behavior" scale. Pediatr Crit Care Med. 2005;6:58‐63.Dorfman TL, Sumamo Schellenberg E, Rempel GR, Scott SD, Hartling L. An evaluation of instruments for scoring physiological and behavioral cues of pain, non‐pain related distress, and adequacy of analgesia and sedation in pediatric mechanically ventilated patients: a systematic review. Int J Nurs Stud. 2014;51:654‐676.Harris J, Ramelet AS, van Dijk M, et al. Clinical recommendations for pain, sedation, withdrawal and delirium assessment in critically ill infants and children: an ESPNIC position statement for healthcare professionals. Intensive Care Med. 2016;42:972‐986.Boerlage AA, Ista E, Duivenvoorden HJ, de Wildt SN, Tibboel D, van Dijk M. The COMFORT behaviour scale detects clinically meaningful effects of analgesic and sedative treatment. Eur J Pain. 2015;19:473‐479.Franck LS, Harris SK, Soetenga DJ, Amling JK, Curley MA. The Withdrawal Assessment Tool‐1 (WAT‐1): an assessment instrument for monitoring opioid and benzodiazepine withdrawal symptoms in pediatric patients. Pediatr Crit Care Med. 2008;9:573‐580.Zaccagnini M, Ataman R, Nonoyama ML. The withdrawal assessment tool to identify iatrogenic withdrawal symptoms in critically ill paediatric patients: a COSMIN systematic review of measurement properties. J Eval Clin Pract. 2021;27:976‐988.Habib E, Almakadma AH, Albarazi M, et al. Iatrogenic withdrawal syndrome in the pediatric cardiac intensive care unit: incidence, risk factors and outcome. J Saudi Heart Assoc. 2021;33:251‐260.Amigoni A, Vettore E, Brugnolaro V, et al. High doses of benzodiazepine predict analgesic and sedative drug withdrawal syndrome in paediatric intensive care patients. Acta Paediatr. 2014;103:e538‐e543.Ávila‐Alzate JA, Gómez‐Salgado J, Romero‐Martín M, Martínez‐Isasi S, Navarro‐Abal Y, Fernández‐García D. Assessment and treatment of the withdrawal syndrome in paediatric intensive care units: systematic review. Medicine (Baltimore). 2020;99:e18502.Bichaff P, Setani KT, Motta EHG, Delgado AF, Carvalho WB, Luglio M. Opioid tapering and weaning protocols in pediatric critical care units: a systematic review. Rev Assoc Med Bras. 1992;2018(64):909‐915.Amirnovin R, Sanchez‐Pinto LN, Okuhara C, et al. Implementation of a risk‐stratified opioid and benzodiazepine weaning protocol in a pediatric cardiac ICU. Pediatr Crit Care Med. 2018;19:1024‐1032.Sanavia E, Mencía S, Lafever SN, Solana MJ, Garcia M, López‐Herce J. Sedative and analgesic drug rotation protocol in critically ill children with prolonged sedation: evaluation of implementation and efficacy to reduce withdrawal syndrome. Pediatr Crit Care Med. 2019;20:1111‐1117.Elyas M, Haggag S, Atef A. 1306 Implementation of Withdrawal Assessment Tool and Weaning Protocol to Reduce Iatrogenic Withdrawal Syndrome in Pediatric Cardiac ICU: a Quality Improvement Project. BMJ Publishing Group Ltd; 2022.Hernán MA, Hernández‐Díaz S. Beyond the intention‐to‐treat in comparative effectiveness research. Clin Trials. 2012;9:48‐55.da Silva PS, Reis ME, Fonseca TS, Fonseca MC. Opioid and benzodiazepine withdrawal syndrome in PICU patients: which risk factors matter? J Addict Med. 2016;10:110‐116.Geslain G, Ponsin P, Lãzãrescu AM, et al. Incidence of iatrogenic withdrawal syndrome and associated factors in surgical pediatric intensive care. Arch Pediatr. 2022;30:14‐19.

Journal

Acta Anaesthesiologica Scandinavica – Wiley

Published: Oct 1, 2023

Keywords: algorithms; analgosedation; iatrogenic withdrawal syndrome; pediatric intensive care

There are no references for this article.