Secondary Logo

Journal Logo

Editorials: Editorial

Postanesthesia Care Unit Costs Are Heterogeneous Among Hospitals, Principally Determined by Delays in Patient Admission From Operating Rooms

Dexter, Franklin MD, PhD, FASA*; Epstein, Richard H. MD, FASA

Author Information
doi: 10.1213/ANE.0000000000004110
  • Free

See Article, p

In the current issue of Anesthesia & Analgesia, Grabitz et al1 examined the effect of postoperative residual neuromuscular blockade on the incidence of unplanned intensive care unit admission and on hospital costs. The authors studied a cohort of patients who underwent general anesthesia including administration of a nondepolarizing neuromuscular blocking agent between 2008 and 2013 at Massachusetts General Hospital. Their independent variable was postoperative residual neuromuscular blockade on admission into the postanesthesia care unit (PACU).1 Their definition of residual neuromuscular blockade was the presence of a train-of-four ratio <0.9, measured using accelerometry.1 The incidence of unplanned intensive care unit admission was greater among patients who were not fully reversed, 3.3% vs 1.5%. The adjusted odds ratio of intensive care unit admission was 3.03, with 95% CI of 1.33–6.87. In addition, they previously reported that in the cohort through 2010, the PACU time was greater in the patients with postoperative residual neuromuscular blockade, mean 323 minutes (SD = 299 minutes) vs 243 minutes (SD = 185 minutes).2 The mean difference between groups of 80 minutes was reasonable for the extra PACU time associated with medical causes.3–5 Nevertheless, the authors report in their current study1 that there was no significant difference in total hospital costs between groups, with the adjusted ratio equaling 1.04, with 95% CI of 0.98–1.11. The authors treated direct variable costs as those varying with the patient’s specific care (eg, medications). The adjusted ratio for direct variable costs was 1.03, with 95% CI of 0.96–1.11.

The authors did not identify reasons why the greater incidence of admission into the intensive care unit and longer times in the PACU was not associated with detectably greater costs.1,2 Thus, precisely what accounted for their findings cannot be determined from their reported data. However, their cost result makes sense, conceptually.

First, the greater odds of intensive care unit admission associated with partial paralysis show concurrent validity of the authors’ finding. Postoperative residual neuromuscular blockade has safety consequences for patients. However, the patients were admitted to the intensive care unit for further monitoring, not additional procedures such as mechanical ventilation.1 Thus, the intensive care unit admission was effectively, for many patients, equivalent to a longer stay in the PACU.1 There would be only small increases in costs for the entire population because the difference in the incidence of intensive care unit admission was only 1.8%.1 The cohort studied was all patients; consequently, the absolute number of admissions is the relevant end point economically. The value is 2 extra patients per year.a That would be too few patients to influence intensive care unit staffing. Furthermore, from the authors’ discussion:

There was a “low ICU bed occupancy in our study center. Lower occupancy facilitates early ICU 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 nursing patient ratio equivalent to the ICU. Effects of residual neuromuscular blockade on costs of care may be different in a clinical scenario where procedures would need to be cancelled as a result of lack of availability of ICU beds.”1

Second, the principal incremental cost from staying longer in the PACU is not the cost of the time spent in the PACU. Rather, the incremental cost is the difference between the extra time spent in the PACU versus the time that otherwise would have been on the hospital ward.5 From the investigators’ previous study,2 consider a mean difference in time in the PACU of 80 minutes. In the phase I PACU before patient transfer to a hospital ward, each nurse generally cares for 2 patients. On a hospital ward, among patients newly admitted, a nurse may care for 4 such patients. We use the national median hourly registered nurse compensation of $33.65 per hour.6 The incremental cost of the PACU time from postoperative residual neuromuscular blockade would be approximately $11.22, where $11.22 = ($33.65 per hour) × (80 minutes/60 minutes per hour) × (1 nurse/2 patients − 1 nurse/4 patients).7 This calculation overstates the savings because some patients staying for several hours would have transitioned to phase II care, where the ratio can be 1 nurse:3 patients. Regardless, small ($11.22) incremental costs for PACU time were to be expected. Even with perioperative patient care and management processes in effect in 1995, elimination of 100% of adverse events among PACU patients, including all pain, respiratory complications, nausea, vomiting, etc, would have reduced total PACU time among all patients by <9.0%.4 A 50% reduction would have reduced PACU nursing time by <4.8%.8 An implication for the authors’ results1 is obtained by comparing the estimate of $11.22 to the authors’ estimated SD of total hospital costs in their Methods section labeled “Power Analysis.” The ratio, being ≈0.03%, shows that the increases (eg, ≈80 minutes) in the time spent in the PACU could not result in substantively detectable increases in costs among patients because of the variability among patients in costs.

Based on the results published this month by Grabitz et al,1 we have a recommendation to readers who are contemplating the hospital cost of longer PACU time from postoperative residual neuromuscular blockade. Focus attention on the implications of delays in patient admission from operating rooms into the PACU, not on the extra PACU time, per se. The effect of the postoperative residual neuromuscular blockade on the percentage of days with at least 1 delay in admission was what the investigators calculated in their earlier study.2,b The extra time spent in the PACU increases the PACU census and can result in a delay in transferring another patient from the operating room into the PACU.3,7,9 The principal economic savings from reducing time in the PACU would be achieved by the subset of hospitals whose PACUs frequently are full.7,9 Because the investigators’ hospital had 55% of days with a delay from the operating rooms, these conditions held.2,7,9 Thus, it would not be the cost of PACU time of the patient with postoperative residual neuromuscular blockade that matters per se, but rather the resulting increase in operating room time for other patients. There has been good understanding for 15 years in how to quantify and value operating room time caused by longer PACU time.3,10 However, that is not the valuation for the individual patient and thus would not be included in most hospitals’ cost accounting other than by a relatively artificial inflation of the cost per minute of PACU time (see below).

The reason the statistical relationship between PACU time and cost is not simply a matter of arithmetic is 3-fold. First, delays from operating room into a PACU are autocorrelated (eg, if 1 patient is delayed on a day at 2:00 pm, there is a greater than typical chance that the next patient admitted will also have a delay in the operating room). Second, studies need to consider information about other patients in the surgical suite at the time of delay who may be admitted to the same PACU, increasing the analytic complexity. Third, whatever is learned about delays at one hospital may not be generalizable to another hospital.11 In other words, even if at a studied hospital there were frequent delays from operating rooms into the PACU, that condition would not apply to some other hospitals, and vice versa. For example, delays from the operating room into the PACU would not apply to Japanese hospitals with most patients planned to recover in operating rooms.12,13 At hospitals with frequent delays and little heterogeneity among operating rooms in total hours of cases, case sequencing can substantively reduce time costs (eg, by 11.8%).14,15 Also, costs depend on the proportion of PACU nurses working part-time and paid hourly.11 The implication is that whereas the mean incremental time spent in the PACU is generalizable, the incremental (large) increase in costs from greater operating room time should be expected to be heterogeneous among hospitals. This is not heterogeneity caused by a lack of scientific understanding or by poor study design. Rather, it is heterogeneity from hospitals being different.

Focusing on delays from the operating rooms into the PACU will not necessarily have substantive financial benefit, but the potential savings will be greater than that for saving PACU time.9 Because operating room time is vastly more expensive than PACU time, most hospitals rarely have delays from operating room into PACU (ie, they adjust the number of PACU beds and their staffing to try to mitigate such occurrences). The benefit of reducing PACU time on preventing delays in admission is conditional on the more important first step of planning available PACU nurses’ shifts based on reducing the days with at least 1 such delay.7,9,10

Our general recommendation to readers is to use the above examples to consider what it means for labor cost in the PACU to be considered as a variable cost. Do not misinterpret cost as being proportional to time spent in the PACU. For a population of many patients, the costs of drugs and supplies are negligible relative to nursing labor costs.3 Thus, it is important to focus on the time patients spend in the PACU, as the investigators did in their previous report of their patient cohort.2 However, one should not model PACU cost as being strictly proportional to time (ie, do not literally apply our arithmetic above which produced an incremental cost of residual paralysis of $11.22 or with a much larger cost per minute value to represent that some patients have delay from operating rooms into PACUs). For example, consider a patient in the PACU who is resting comfortably. The patient’s next stop will be transfer to a hospital ward. There are currently several empty PACU beds, enough for all patients expected to be admitted from the operating rooms during the next 1 hour.16 Having the patient remain in the PACU for an extra 40 minutes (using half that observed and typical) does not literally cost ≤$5.11 (see above), it costs $0. The reason is that keeping the patient in the PACU 40 minutes longer has not changed the PACU nurse staffing (ie, total numbers present working different hours) or staff scheduling (ie, which nurses are working which shifts).16 The PACU nurses and ward nurses will be present regardless of whether the patient stays in the PACU for an extra 40 minutes. Perhaps with the closer nursing attention in the PACU, the patient may have somewhat greater comfort (ie, staying longer will have benefit). Accounting of PACU time as being proportional to time in the PACU is reasonable for annual decision making (eg, for purposes of comparing annual costs among facilities). However, one should not misapply such accounting to the cost of care of individual patients.


Name: Franklin Dexter, MD, PhD, FASA.

Contribution: This author helped write the manuscript.

Name: Richard H. Epstein, MD, FASA.

Contribution: This author helped write the manuscript.

This manuscript was handled by: Ken B. Johnson, MD.


aThe value of 2.0 = (15 patients with unplanned admission in the group with postoperative residual neuromuscular blockade) − (457 patients in that group) × (27 patients with unplanned admission in the other group)/(1776 patients in the other group).

b“Assuming that no patient had postoperative residual curarization, the probability of at least 1 patient waiting to enter the PACU during the day is 55%. If all patients experienced postoperative residual curarization and had the same prolongation of their stay as observed in our sample, this probability would increase to 95%.”2


1. Grabitz SD, Rajaratnam N, Changani K, et al. The effects of postoperative residual neuromuscular blockade on hospital costs and intensive care unit admission: a population-based cohort study. Anesth Analg. 2019;128:1129–1136.
2. 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.
3. Dexter F, Tinker JH. Analysis of strategies to decrease postanesthesia care unit costs. Anesthesiology. 1995;82:94–101.
4. Chung F, Mezei G. Factors contributing to a prolonged stay after ambulatory surgery. Anesth Analg. 1999;89:1352–1359.
5. Dexter F, Penning DH, Traub RD. Statistical analysis by Monte-Carlo simulation of the impact of administrative and medical delays in discharge from the postanesthesia care unit on total patient care hours. Anesth Analg. 2001;92:1222–1225.
6. Bureau of Labor Statistics. Occupational outlook handbook, registered nurses. Available at Accessed December 24, 2018.
7. Dexter F, Wachtel RE, Epstein RH. Impact of average patient acuity on staffing of the phase I PACU. J Perianesth Nurs. 2006;21:303–310.
8. Cohen MM, O’Brien-Pallas LL, Copplestone C, Wall R, Porter J, Rose DK. Nursing workload associated with adverse events in the postanesthesia care unit. Anesthesiology. 1999;91:1882–1890.
9. Dexter F, Epstein RH, Marcon E, de Matta R. Strategies to reduce delays in admission into a postanesthesia care unit from operating rooms. J Perianesth Nurs. 2005;20:92–102.
10. Dexter F, Epstein RH, Penning DH. Statistical analysis of postanesthesia care unit staffing at a surgical suite with frequent delays in admission from the operating room–a case study. Anesth Analg. 2001;92:947–949.
11. Dexter F, Macario A, Manberg PJ, Lubarsky DA. Computer simulation to determine how rapid anesthetic recovery protocols to decrease the time for emergence or increase the phase I postanesthesia care unit bypass rate affect staffing of an ambulatory surgery center. Anesth Analg. 1999;88:1053–1063.
12. Thenuwara KN, Yoshi T, Nakata Y, Dexter F. Time to recovery after general anesthesia at hospitals with and without a phase I post-anesthesia care unit: a historical cohort study. Can J Anesth. 2019;12:1296–1302.
13. Sento Y, Suzuki T, Suzuki Y, Scott DA, Sobue K. The past, present and future of the postanesthesia care unit (PACU) in Japan. J Anesth. 2017;31:601–607.
14. Marcon E, Dexter F. An observational study of surgeons’ sequencing of cases and its impact on postanesthesia care unit and holding area staffing requirements at hospitals. Anesth Analg. 2007;105:119–126.
15. Bai M, Storer RH, Tonkay GL. A sample-gradient-based algorithm for a multiple-OR and PACU surgery scheduling problem. IIE Trans. 2017;49:367–380.
16. Ehrenfeld JM, Dexter F, Rothman BS, et al. Lack of utility of a decision support system to mitigate delays in admission from the operating room to the postanesthesia care unit. Anesth Analg. 2013;117:1444–1452.
Copyright © 2019 International Anesthesia Research Society