Among the 3000 patients in the database who underwent general anesthesia and who had train-of-four ratio assessment in the PACU, 2233 met inclusion criteria for our primary analysis (Figure). There were 457 (20.5%) patients diagnosed with postoperative residual neuromuscular blockade in the PACU. Proportion of males and mean Charlson comorbidity index were higher in patients experiencing postoperative residual neuromuscular blockade (Table 1). The majority of the study patients were admitted on the day of surgery (n = 1930).
Hospital Costs (Primary Analysis)
On initial unadjusted truncated negative binomial regression analysis, postoperative residual neuromuscular blockade was associated with an increase in total hospital costs (incidence rate ratio, 1.14, 95% CI, 1.06–1.22; P < .001). Following adjustment for our a priori defined confounders, there was no longer a significant difference in hospital costs between patients experiencing postoperative residual neuromuscular blockade and those without residual paralysis (adjusted incidence rate ratio, 1.04, CI, 0.98–1.11; P = .22). The findings from the primary regression analysis remained robust using the imputed database (adjusted incidence rate ratio, 1.05, CI, 0.99–1.12; Table 2). Further assessment of effect modification as suggested by the editor and reviewers is provided in Supplemental Digital Content, Document, Section 4, http://links.lww.com/AA/C698.
Intensive Care Unit Admission and Hospital Length of Stay (Secondary Analysis)
A total of 42 patients of n = 2233 were postoperatively admitted to the intensive care unit, giving an incidence rate of 1.8%. The incidence rates in the postoperative residual neuromuscular blockade and no postoperative residual neuromuscular blockade group were 3.3% (n = 15) and 1.5% (n = 27), respectively. There were no fatalities in this cohort. The adjusted odds of intensive care unit admission in the postoperative residual neuromuscular blockade group compared to the unexposeed group was 3.03, (CI, 1.33–6.87; P = .008). The adjusted incidence rate ratio of hospital length of stay was 1.09 (CI, 1.00–1.19; P = .058) in the postoperative residual neuromuscular blockade group (Table 2).
Although postoperative residual neuromuscular blockade was defined as a train-of-four ratio <0.9 in this study and in other similar studies,3,9 we further explored the connection between postoperative residual neuromuscular blockade and our primary outcome, total hospital costs, in patients with postoperative train-of-four ratio <0.8 and train-of-four ratio <0.7, which indicate a deeper level of residual neuromuscular blockade.9 The adjusted analyses showed no significant differences between the exposure and control groups: train-of-four <0.8 (adjusted incidence rate ratio, 1.05, CI, 0.94–1.17; P = .35) and train-of-four <0.7 (adjusted incidence rate ratio, 1.08, CI, 0.86–1.36; P = .49). Thus, in our cohort, a correlation between postoperative residual neuromuscular blockade and costs of the index hospitalization was not observed even at increased levels of postoperative residual neuromuscular blockade.
We repeated our costs analysis using the components of total costs as the dependent variables. As highlighted in Table 2, the unadjusted incidence rate ratio for variable direct costs was 1.15 (CI, 1.07–1.12) and the adjusted incident rate ratio was 1.03 (CI, 0.96–1.11). We performed sensitivity analysis using the subgroup day of surgery admissions and ambulatory surgery to exclude indirect costs of prior hospital admission up to the event “postoperative residual neuromuscular blockade in PACU.” In the day of surgery admission subgroup, the adjusted incident rate ratio for increased costs was 1.04 (CI, 0.99–1.09).
When additionally adjusting the secondary regression model for preoperative opioid use, we found a significant association between postoperative residual neuromuscular blockade and higher odds of postoperative intensive care unit admission. Including the opioid prescription on the day of discharge as a covariate in this regression analysis again confirmed robustness of the association between postoperative residual neuromuscular blockade and postoperative intensive care unit admission.
A variable indicating if the surgical case was completed in the morning (7 AM–11:59 AM), in the afternoon (noon – 4:59 PM), or at night (5 PM–6:59 AM) was included as covariate in an additional sensitivity analysis, which confirmed the independent association of postoperative residual neuromuscular blockade and postoperative admission to the intensive care unit (Table 2).
In this retrospective study of 2233 patients who underwent general anesthesia, 457 (20.5%) demonstrated postoperative residual neuromuscular blockade on admission to the PACU. Our analyses revealed that postoperative residual neuromuscular blockade was not associated with a significant increase in estimated total or direct variable health care costs. However, we found that postoperative residual neuromuscular blockade in the PACU was associated with a 3-fold increase in the odds of being admitted to the intensive care unit.
Consistent with a previously published incidence rate of 22% of postoperative residual neuromuscular blockade at our institution,4 we observed an incidence rate of 20.5% of postoperative residual neuromuscular blockade in this cohort. Based on the original protocol, patients within the original study were consecutively screened for recruitment if they received general anesthesia with nondepolarizing neuromuscular blocking agent. The short, predefined time frame of the train-of-four assessment (10 minutes after PACU admission) and the high rate of consecutively enrolled patients (96%) minimized selection bias.
Rates of postoperative residual neuromuscular blockade in other studies vary from 4% to 64%.3–8 However, these studies did not report reintubations or admissions to the intensive care unit. We previously reported an associated increase in the mean PACU length of stay of 80 minutes with postoperative residual neuromuscular blockade (P = .03).4 In 2012, Thilen et al5 studied postoperative residual neuromuscular blockade in 150 patients and reported postoperative residual neuromuscular blockade incidence rate of 52% in their group via adductor pollicis muscle monitoring. Although some patients (n = 13) in that cohort were electively ventilated in the PACU to facilitate regional anesthesia, there were no reported respiratory complications or intensive care unit admissions. The Residual Curarization and its Incidence at Tracheal Extubation study reported one of the highest incidences of postoperative residual neuromuscular blockade across all studies (63.5%), when the phenomenon was screened for in 302 postabdominal surgery patients across 8 Canadian hospitals, with only 1 patient requiring reintubation. This study did not report intensive care unit admission data or costs.3 Despite these high incidence rates, routine monitoring to evaluate the reversal of postoperative residual neuromuscular blockade is not considered part of minimum monitoring standards in most clinical settings. A large international survey in 2010 revealed that 19.3% of European and 9.4% of US clinicians do not use neuromuscular monitors in postoperative evaluations and that pharmacological reversal was routinely administered by only 18% of European and 34.2% of US clinicians surveyed.7
Our secondary analysis demonstrated a 3-fold increase in the odds of intensive care unit admission from the PACU in those determined to have postoperative residual neuromuscular blockade. This finding has not been previously reported. Previous studies have shown an association between postoperative residual neuromuscular blockade and both increased respiratory complications and delayed discharge from the PACU,1,4 but few have reported unanticipated intensive care unit admission rates. For instance, in 2015, McLean et al1 demonstrated a dose-dependent association between intermediate-acting neuromuscular blocking agents and postoperative respiratory complications. Over 48,000 cases were included for analysis of the association between dosing of nondepolarizing neuromuscular blocking agent and a composite outcome of respiratory complications within 3 postoperative days. In this study, the reintubation rate was 0.3% and they did not report intensive care unit admission or hospital costs.
Much debate exists within the literature as to whether intensive care unit admission universally translates into better outcomes, despite presumed increased hospital costs.14,15 Our group has postulated that an acuity threshold may exist below which the risks of intensive care unit admission outweigh the benefits.16 In this study, we found a trend toward increased hospital length of stay in patients with postoperative residual neuromuscular blockade, which in part may be explained by admission of patients with lower acuity except for residual neuromuscular blockade. In contrast to a previous study,17 we found differential effects of residual neuromuscular blockade on intensive care unit admission rate. This may be in part explained by low intensive care unit bed occupancy in our study center. Lower occupancy facilitates early intensive care unit admission, as opposed to keeping patients who need extended postoperative care longer in the PACU. The PACU is a very cost-intensive location in the hospital with a nurse-to-patient ratio equivalent to the intensive care unit. Effects of residual neuromuscular blockade on costs of care may be different in a clinical scenario where procedures would need to be canceled as a result of lack of availability of intensive care unit beds. Therefore, based on the results of our analyses, it would be advisable for clinicians to screen for residual neuromuscular blockade in the perioperative setting before transfer to the PACU, and to take appropriate measures ensuring complete recovery of normal neuromuscular physiology to offset the risk of unnecessary intensive care unit admission or increased PACU length of stay. As our study was conducted in a well-resourced academic center with critical care bed elasticity, further studies to validate our findings using large heterogeneous datasets may provide more insight into the problem of postoperative residual neuromuscular blockade in the PACU and its burden on a variety of intensive care unit structures. Our group recently studied the effect of incentivized protocols for checking adequate reversal of nondepolarizing neuromuscular blocking agent in the perioperative setting and found lower odds of postoperative pulmonary complications, lower costs, and shorter duration of hospital stay after implementation of the quality improvement initiative.18
A major strength of this study is that we measured train-of-four ratio at PACU admission prospectively in a large sample size using industry standard TOFwatch technology (TOF-Watch, Organon, Finland). Our hypothesis was further tested with exploratory analyses using lower train-of-four thresholds for our exposure variable of postoperative residual neuromuscular blockade and still failed to show significantly increased hospital costs. Several sensitivity analyses confirmed the robustness of our findings. We believe our study has face validity, as characterized by (1) the good coverage of surgical patients from Massachusetts General Hospital without preselection of patients causing selection bias; (2) a confounder model that addresses the wide range of comorbidities and procedures; and (3) the absence of cost differences and the minimal beta error observed in our analysis.19,20 The model results are a good representation of the actual cost. For example, the model predicted increase cost of 27% for patients with a high ASA physical status classification (≥III) compared to patients with a lower ASA physical status (I or II). Similarly, the model predicted a cost increase of 87% moving from lowest to highest costly surgical service. Finally, the incidence rate of postoperative residual neuromuscular blockade in the entire cohort was 20.5%, which we believe to be accurate based on the fact that 96% of consecutively recruited patients in the original study had valid train-of-four values at PACU admission.
Our single center, large tertiary referral hospital may not allow generalizability to smaller, less well-resourced health care settings. In addition, despite a robust confounder model, we present observational data where unknown factors may confound the results. Our institution is not able to separate costs before and after PACU admission, so our analysis is limited by data points collected before the exposure, postoperative residual neuromuscular blockade in PACU, which may affect the direct association of postoperative residual neuromuscular blockade on hospital costs.
Perioperative cost data are often dominated by a few outlier patients, which make it hard to identify associations between preventable complications, such as postoperative residual neuromuscular blockade, and costs. One such outlier in our cohort was dependent on home oxygen therapy. Given the highly skewed distribution of costs in our cohort, we conducted a sensitivity analysis excluding this 1 true outlier patient. The main findings did not change when excluding this patient from the analysis; we again found no association of postoperative residual neuromuscular blockade and costs in the adjusted analysis (adjusted incidence rate ratio, 1.03, CI, 0.97–1.10), whereas postoperative residual neuromuscular blockade was significantly associated with postoperative intensive care unit admission. Finally, information about some outcomes (hospital length of stay and costs) was retrieved from administrative data where misclassification is possible.
We found that postoperative residual neuromuscular blockade in the PACU was not associated with increased health care costs, but with a significant increase in the odds of intensive care unit admission. Residual neuromuscular blockade is prevalent and underdiagnosed, and adequate prevention may decrease rates of unplanned intensive care unit admission associated with residual neuromuscular blockade.
Name: Stephanie D. Grabitz, MD.
Contribution: This author helped with literature review, data extraction, coding, study design, data analysis, writing and revision of the manuscript.
Name: Nishan Rajaratnam, MD.
Contribution: This author helped with literature review, data extraction, coding, data analysis, study design, statistical analysis plan, and writing the manuscript.
Name: Khushi Chhagani, BS.
Contribution: This author helped with data extraction, coding, data analysis, study design, statistical analysis plan, and writing the manuscript.
Name: Tharusan Thevathasan, Cand Med.
Contribution: This author helped with literature review, data extraction, coding, study design, and data analysis.
Name: Bijan J. Teja, MD, MBA.
Contribution: This author helped with study design, statistical analysis plan, data analysis, and writing the manuscript.
Name: Hao Deng, MD, MPH.
Contribution: This author helped with study design and statistical analysis planning.
Name: Matthias Eikermann, MD, PhD.
Contribution: This author conceived the study hypothesis, helped with study design, and approved the final manuscript.
Name: Barry J. Kelly, MD, MSc.
Contribution: This author helped with literature review, study design, statistical analysis plan, data analysis, writing, submission and revision of the manuscript.
This manuscript was handled by: Ken B. Johnson, MD.
1. McLean DJ, Diaz-Gil D, Farhan HN, Ladha KS, Kurth T, Eikermann M. Dose-dependent association between intermediate-acting neuromuscular-blocking agents and postoperative respiratory complications. Anesthesiology. 2015;122:1201–1213.
2. Thevathasan T, Shih SL, Safavi KC, et al. Association between intraoperative non-depolarising neuromuscular blocking agent dose and 30-day readmission after abdominal surgery. Br J Anaesth. 2017;119:595–605.
3. Fortier LP, McKeen D, Turner K, et al. The RECITE study: a Canadian Prospective, Multicenter Study of the incidence and severity of residual neuromuscular blockade. Anesth Analg. 2015;121:366–372.
4. Butterly A, Bittner EA, George E, Sandberg WS, Eikermann M, Schmidt U. Postoperative residual curarization from intermediate-acting neuromuscular blocking agents delays recovery room discharge. Br J Anaesth. 2010;105:304–309.
5. Thilen SR, Hansen BE, Ramaiah R, Kent CD, Treggiari MM, Bhananker SM. Intraoperative neuromuscular monitoring site and residual paralysis. Survey Anesthesiol. 2013;57:156–157.
6. Donati F. Residual paralysis: a real problem or did we invent a new disease? Can J Anaesth. 2013;60:714–729.
7. Naguib M, Kopman AF, Lien CA, Hunter JM, Lopez A, Brull SJ. A survey of current management of neuromuscular block in the United States and Europe. Anesth Analg. 2010;111:110–119.
8. Kumar GV, Nair AP, Murthy HS, Jalaja KR, Ramachandra K, Parameshwara G. Residual neuromuscular blockade affects postoperative pulmonary function. Anesthesiology. 2012;117:1234–1244.
9. Murphy GS, Brull SJ. Residual neuromuscular block: lessons unlearned. Part I: definitions, incidence, and adverse physiologic effects of residual neuromuscular block. Anesth Analg. 2010;111:120–128.
10. Sasaki N, Meyer MJ, Malviya SA, et al. Effects of neostigmine reversal of nondepolarizing neuromuscular blocking agents on postoperative respiratory outcomes: a prospective study. Anesthesiology. 2014;121:959–968.
11. Ladha K, Vidal Melo MF, McLean DJ, et al. Intraoperative protective mechanical ventilation and risk of postoperative respiratory complications: hospital based registry study. BMJ. 2015;351:h3646.
12. Brueckmann B, Villa-Uribe JL, Bateman BT, et al. Development and validation of a score for prediction of postoperative respiratory complications. Anesthesiology. 2013;118:1276–1285.
13. Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43:1130–1139.
14. Wunsch H, Gershengorn HB, Cooke CR, et al. Use of intensive care services for Medicare beneficiaries undergoing major surgical procedures. Anesthesiology. 2016;124:899–907.
15. Kahan BC, Koulenti D, Arvaniti K, et al; The International Surgical Outcomes Study (ISOS) Group Critical care admission following elective surgery was not associated with survival benefit: prospective analysis of data from 27 countries. Intensive Care Med. 2017;43:971–979.
16. Thevathasan T, Copeland CC, Long DR, et al. The impact of postoperative intensive care unit admission on postoperative hospital length of stay and costs: a prespecified propensity-matched cohort study. Anesth Analg. 2018 [Epub ahead of print].
17. Dexter F, Blake JT, Penning DH, Sloan B, Chung P, Lubarsky DA. Use of linear programming to estimate impact of changes in a hospital’s operating room time allocation on perioperative variable costs. Anesthesiology. 2002;96:718–724.
18. Rudolph MI, Chitilian HV, Ng PY, et al. Implementation of a new strategy to improve the peri-operative management of neuromuscular blockade and its effects on postoperative pulmonary complications. Anaesthesia. 2018;73:1067–1078.
19. O Neill L, Dexter F. Tactical increases in operating room block time based on financial data and market growth estimates from data envelopment analysis. Anesth Analg. 2007;104:355–368.
20. Dexter F, Epstein RH. Typical savings from each minute reduction in tardy first case of the day starts. Anesth Analg. 2009;108:1262–1267.
Supplemental Digital Content
© 2019 International Anesthesia Research Society