Intraoperative hypothermia may increase the incidences of wound infection,1 1–4 5 1 6 , 7 8 9 10
As part of its response to the proposed rule by CMS regarding the physician fee schedule,11 12 13
In this informatics study, we used anesthesia information management systems (AIMS) data from 4 large, academic hospitals to provide guidance to organizations desiring to report on PQRS #424 or NQF 2681. The primary issue relates to the requirement that the intraoperative temperature is reportable as a quality measure only if it is recorded within 30 minutes before the anesthesia end time. In a study of 50 patients undergoing laparoscopic surgery, the average time from temperature probe removal in the operating room (OR) to postanesthesia care unit (PACU) arrival was 30 ± 12 (SD) minutes.14 15 14
METHODS
The protocol was reviewed by the Institutional Review Board at the University of Miami Hospital (UMH), Thomas Jefferson University Hospital (TJUH), Jackson Memorial Hospital (JMH), and the University of California Los Angeles (UCLA), each of which concluded that this study did not meet the US regulatory definition of human subjects research. De-identified retrospective data were extracted from the AIMS databases at the 4 institutions. The different date ranges for each hospital correspond to the available data in their respective data warehouses at the time of analysis (Table 1 ). None of the 4 hospitals were reporting on PQRS #424 or NQF 2681 during the study interval, no feedback related to intervals from temperature monitoring discontinuation to the anesthesia end time was provided, and no instructions were transmitted to anesthesia providers related to maintaining intraoperative measurement of temperature as long as possible. The Standards for Quality Improvement Reporting Excellence (SQUIRE) checklist was followed in the preparation of this manuscript.16
Table 1.: Summary of Study Datasets and Cases Excluded Due to Missing or Implausible Values
Similar to the Sun et al17 17
For each case, all sequentially recorded temperatures (nominally at 1-minute intervals) were retrieved. We defined the “end-of-case temperature” as the maximum intraoperative temperature within 30 minutes before the anesthesia end time; this would be the intraoperative temperature used for quality reporting purposes. Intervals were calculated between the anesthesia end time and the following intraoperative milestone events: (1) the time of the last intraoperative temperature (“last temperature time”); (2) the time when surgery end occurred (“surgery end”); and (3) the time of extubation or removal of a supraglottic airway (“extubation”). We analyzed the interval from both end of surgery and extubation to the anesthesia end time because these represent 2 potential strategies for groups wishing to report temperature using only intraoperative measurements. That is, providers would be instructed to maintain core temperature recording until one of these events occurred, rather than to discontinue monitoring sooner.
PACU temperatures were available electronically at UCLA, but not at the other 3 hospitals. These PACU temperatures were measured from a variety of sites, including oral, temporal artery, bladder, axillary, and tympanic membrane. Because the national quality measures neither distinguish among temperature sources, nor provide different thresholds for core or surface measurements, PACU temperature data from the various sites were pooled. No clinical information was retrieved related to either the patients or other aspects of their surgery.
Statistical Methods
At each hospital, using all cases, the intervals >0 minutes from the 3 milestone events (“last temperature time,” “end surgery,” “extubation”) to the anesthesia end event were sorted. The fractions of cases where the interval was >30 minutes were determined. The 95% Clopper–Pearson binomial confidence intervals (CIs) for these fractions were calculated. These values represent the percentages of cases for which the intraoperative temperature would have been ineligible for reporting (see Table 2 ). The fractions of cases excluded from reporting using the last temperature time were conservative (ie, an underestimate), because the time of the last temperature would always be the same as or later than the time of the maximum temperature. Calculations were performed using Excel formulas (Microsoft, Redmond, WA).
Table 2.: Percentages of Case Where the Intraoperative Temperature Would Be Ineligible for Reporting Purposes Because Interval to the Anesthesia End Time Was >30 Minutes
At each hospital, for each quarter, the mean of the end-of-case temperature and the fraction of cases where this temperature was <35.5°C were calculated. The paired Student t test was used to compare these differences to 0 using Systat 13 (Systat Software, Inc, San Jose, CA), with P < .05 necessary to claim statistical significance. These values represent the percentage of cases that would have failed the performance measure threshold of 35.5°C (see Table 3 ).
Table 3.: End-of-Case Intraoperative Temperaturesa
In the Appendix, using data from UCLA, temperatures from both the OR and the PACU were considered for all cases meeting the PQRS #424 reporting requirements (ie, all patients, regardless of age, having general or neuraxial anesthesia with a case duration ≥60 minutes). The objective of this observational analysis was to determine the extent to which cases from this institution would have satisfied the national reporting measure. This process was not feasible for the other study hospitals because of the absence of electronic temperature data in the PACU.
No power analysis was conducted because all qualifying cases, comprising between 12 and 30 quarters of data at each hospital, were analyzed. Thus, we expected tiny standard errors. The date ranges for each institution corresponded to the availability of electronically recorded temperature measurements in their respective AIMS databases.
RESULTS
The numbers of cases and data intervals available for analysis (Table 1 ) were as follows: TJUH (84,283, 28 quarters); UMH (29,020, 12 quarters); JMH (65,543, 29 quarters); and UCLA (56,680, 13 quarters). Numbers of cases excluded and corresponding explanations are presented in Table 1 .
Calculations of the percentages of cases that would be excluded from reporting under PQRS #424 due to an interval >30 minutes from various intraoperative milestones are presented in Table 2 . Based on when intraoperative temperature monitoring was discontinued by the anesthesia provider, averages (binned by quarters) of 34.5% to 59.5% of all cases would have been ineligible to use an intraoperative temperature because the interval from the last temperature until anesthesia end was >30 minutes (Table 2 ).a Table 2 ). Figure 1 shows the interval data from UMH from each of the 3 milestones. Corresponding graphs from the other hospitals studied are provided in Supplemental Digital Content, Document, https://links.lww.com/AA/B746
Figure 1.: Distribution of intervals from various anesthesia milestone events until the anesthesia end time. This graph corresponds to the data in Figure 3. Data are presented from the University of Miami Hospital, and are representative of similar graphs at the other study hospitals (see Supplemental Digital Content, Document,
https://links.lww.com/AA/B746 ). Distributions are right skewed, and have been truncated at 120 minutes for display purposes. In Panel A, the intervals from when temperature monitoring was discontinued until the anesthesia end time are shown. To qualify for inclusion in PQRS #424 or NQF 2861 reporting, the interval until the anesthesia end time must be ≤30 minutes. In Panel B, the intervals from when the end of surgery was documented until the anesthesia end time are shown. In Panel C, the intervals from when the tracheal tube or supraglottic airway was removed (“Extubation”) until the anesthesia end time are shown. Figures from all 4 study hospitals are available as Supplemental Digital Content, Document,
https://links.lww.com/AA/B746 .
There was heterogeneity among the 4 study hospitals in the end-of-case temperature (ie, maximum temperature ≤30 minutes before the anesthesia end time), ranging from 36.1°C to 36.5°C (Table 3 ), but each was close to the final core (esophageal) temperature reported by Sun et al (36.3°C).18 Table 3 ). In practice, for such patients, obtaining a PACU temperature ≥35.5°C up to 15 minutes after the anesthesia end time would be required to meet the temperature quality measure. At UCLA, among cases with both an end-of-case temperature and PACU temperature eligible for quality reporting, only 1.2% (N = 13 quarters; 95% CI, 1.0%–1.4%) of the PACU temperatures were <35.5°C, whereas 16.4% (N = 13; 95% CI, 15.3%–17.5%) of the OR values were below this threshold (P < .0001 by paired Student t test).
In the Appendix, we provide details of an algorithm to generate a PQRS or NQF perioperative temperature report and results from its application using OR and PACU data from UCLA. The Appendix highlights the issues related to missing temperatures and artifacts, presence of intraoperative temperatures from multiple sites, and use of PACU temperatures. The Appendix provides guidance to departments who will need to present a detailed data specification to the reporting group within their hospital’s information systems department.
DISCUSSION
Our results demonstrate that for 34.5% to 59.5% of cases among the 4 hospitals studied, the end-of-case (ie, intraoperative) temperature could not be used for PQRS #424 or NQF 2681 reporting because the interval between cessation of temperature monitoring in the OR to the anesthesia end time exceeded 30 minutes. Even if temperatures had been recorded until the time of extubation, between 5.9% and 20.8% of cases would have been excluded solely based on timing considerations. Because the mean percentage of cases for which process data were reported under PQRS #193 in 2013 was 93.11%,19
To compare our temperature results to Sun et al17
The example from UCLA in the Appendix highlights the challenges of developing a scalable, automated process for even as “simple” a measurement as the final intraoperative temperature. Information system analysts responsible for generating such reports will need a data specification that indicates, in detail, how to deal with the various data problems noted. These challenges are likely generalizable to all physiologic parameters that are potential sources for automated quality reporting.
The ASA application to the NQF requesting incorporation of PQRS #424 as a quality measurement stated “… it has become possible for an attentive anesthesia provider to maintain normothermia in any patient under general anesthesia, regardless of surgical conditions, and this is now the expected standard of care [emphases added].”20 21
There are several limitations to our study. First, all 4 study hospitals were large, academic institutions, with many cases performed by trainees. Thus, some of the time intervals between case milestones and the anesthesia end time might be longer than in nonacademic practice settings, reducing the number of manually retrieved PACU temperatures that would be needed. However, it is not likely that these intervals are considerably longer. At one of the study hospitals (TJUH), it was previously reported that the time from extubation to exiting the OR was ≤15 minutes in 84.6% ± 0.4% (standard error) of cases,22 23
Second, there is much greater penetrance of AIMS in academic settings24 18 25–27
A third limitation is that the sites of intraoperative temperature monitoring were generally not documented. However, this makes our data more reflective of the real-world application of PQRS #424 and NQF 2681 because these measures do not allow for the exclusion of cases based on the site of measurement. From the perspective of CMS regulation, it is irrelevant how accurately or from what site temperatures are measured. Rather, all that matters are the values and that they were measured within the allowable time frame. For hospitals desiring to report on national perioperative temperature measures, this may result in focused attention on more accurate capture of temperature and/or not prematurely discontinuing measurement in the OR. This may also include the application of noninvasive technologies that better reflect core values, such as those based on thermal flux using cutaneous probes.28
A fourth limitation is that we excluded neuraxial cases from study. We did this because we wanted to focus, as much as possible, on core temperature measurement. For most neuraxial cases, temperatures are measured from skin sites. Including these cases would have created a negative bias for the end-of-case intraoperative temperatures, as skin temperature is considerably lower than core temperature. (For patients with an intraoperative temperature below the threshold, checking the arrival PACU temperature would be desirable to determine if the measure passed.) The use of noninvasive core temperature devices (see previous paragraph) would eliminate such bias. There would seldom be delays between the end of surgery and leaving the OR among patients receiving a neuraxial anesthesia, in contrast to patients undergoing general anesthesia with prolonged emergence.29–31
A fifth limitation is that we excluded children in our primary analyses, as did Sun et al.18
A sixth limitation of our study’s relevance is that it relates only to the specifications of PQRS #424 and NQF 2681 as they are written and would be used in practice, not to the science of temperature monitoring and intraoperative thermoregulation. First, when temperature was measured continuously, the presence (versus absence) of active warming decreased morbidity (ie, the demonstrations of improved outcomes were not based on the end-of-case temperatures, but rather on the group assignment to the use or disuse of active warming).1–5 32 33 18 18 1 , 2 , 5 34 28
Figure 2.: Flow diagram describing the sequential steps for processing temperatures for national quality reporting. Qualifying cases are all procedures lasting at least 60 minutes. The anesthesia end time is the billing time on the professional fee invoice. For temperatures to qualify, they must be measured during the interval from 30 minutes before the anesthesia end time to within 15 minutes after the anesthesia end time. The measure fails if no temperature is present in the specified interval, or if the maximum temperature during the reporting interval including both OR and PACU temperatures is <35.5°C. OR, operating room; PACU, postanesthesia care unit.
In summary, we found that successful reporting for the perioperative temperature quality measures PQRS #424 or NQF 2681 will likely depend on the ability to capture PACU temperatures. For the many hospitals where PACU care is documented on paper, substantive manual effort will be required. A substantial rate of failure to meet the threshold temperature of 35.5°C should be anticipated if using intraoperative values alone, obtained with measurement technologies currently in widespread use and current patterns of discontinuation of temperature monitoring (ie, before the end of surgery or at extubation). Hospitals considering reporting on perioperative temperature maintenance measures to satisfy anesthesia quality standards should consider the findings in this article when establishing their temperature quality program. For high compliance reporting with minimal additional effort on the part of clinicians, automated methods of data capture and use of highly accurate temperature measurement technologies (ie, reflective of core temperature) that can be applied noninvasively (ie, applicable for all patients) and used continuously during the entire perioperative period likely will be useful.
Appendix. Algorithm to Process Intraoperative Temperatures for Quality Reporting at UCLA
Algorithm Details (see Figure 2 )
Select cases qualifying for temperature reporting
Exclude cases with interval from Enter OR to Leave OR <60 minutes
Exclude cases neither GA nor neuraxial anesthesia
) UCLA: PACU handoff note contains the anesthesia type of general or neuraxial
) Hospitals where the anesthesia type is not reliably noted could exclude cases not GA or neuraxial as:
Not intubated and no tracheal tube and no supraglottic airway device noted
No volatile agent >0.5 minimum alveolar concentration (MAC)
No note in the anesthesia record indicating either spinal or epidural
Extract temperature data with patient location (OR or PACU), timestamp and site of measurement within 30 minutes before and 15 minutes after the anesthesia end time
Remove temperatures >40°C as likely artifacts
If temperatures recorded from multiple sites, order them by preference for use (eg, esophageal before skin temperature)
Use the temperature from the 1° source if present, then 2° source if absent, etc, else null if no temperature for the minute
If no temperatures are present for a case, set the temperature = null
Determine the maximum intraoperative temperature for each case
End-of-case temperature
Determine if the end-of-case OR temperature is ≥35.5°C
If yes, measure passes
Determine the maximum PACU temperature
Is the maximum PACU temperature ≥35.5°C
If yes, measure passes
If no, measure fails
Data from UCLA Between July 1, 2015 and June 30, 2016
Compliance rate = # cases passing/ (# cases passing + # cases failing)
Summary
Among all cases requiring PQRS temperature:
46.4% would pass using the end-of-case OR temperature alone
84.1% would pass using the maximum temperature from 30 minutes before to 15 minutes after the anesthesia end time
15.9% of cases would fail the temperature measure
DISCLOSURES
Name: Richard H. Epstein, MD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Name: Franklin Dexter, MD, PhD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Name: Ira S. Hofer, MD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Name: Luis I. Rodriguez, MD.
Contribution: This author helped conduct the study and write the manuscript.
Name: Eric S. Schwenk, MD.
Contribution: This author helped conduct the study and write the manuscript.
Name: Joni M. Maga, MD.
Contribution: This author helped conduct the study and write the manuscript.
Name: Bradley J. Hindman, MD.
Contribution: This author helped write the manuscript.
This manuscript was handled by: Richard C. Prielipp, MD.
REFERENCES
1. Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. Study of Wound Infection and Temperature Group. N Engl J Med. 1996;334:1209–1215.
2. Schmied H, Kurz A, Sessler DI, Kozek S, Reiter A. Mild hypothermia increases blood loss and transfusion requirements during total hip arthroplasty. Lancet.1996;347:289–292.
3. Winkler M, Akça O, Birkenberg B, et al. Aggressive warming reduces blood loss during hip arthroplasty. Anesth Analg. 2000;91:978–984.
4. Rajagopalan S, Mascha E, Na J, Sessler DI. The effects of mild perioperative hypothermia on blood loss and transfusion requirement. Anesthesiology. 2008;108:71–77.
5. Frank SM, Fleisher LA, Breslow MJ, et al. Perioperative maintenance of normothermia reduces the incidence of morbid cardiac events. A randomized clinical trial. JAMA. 1997;277:1127–1134.
7. Centers for Medicare and Medicaid Services. Medicare and Medicaid Programs; Electronic Health Record Incentive Program-Stage 3 and Modifications to Meaningful Use in 2015 through 2017. Available at:
https://www.federalregister.gov/documents/2015/10/16/2015–25595/medicare-and-medicaid-programs-electronic-health-record-incentive-program-stage-3-and-modifications . Accessed February 13, 2017.
9. Centers for Medicare and Medicaid Services. PQRS Temperature Measure 424. Available at:
https://pqrs.cms.gov/dataset/2016-PQRS-Measure-424-11-17-2015/5mqt-5mnm/data . Accessed February 13, 2017.
10. National Quality Forum. NQF-Endorsed Measures for Surgical Procedures, 2015. Final Report. Available at:
http://www.qualityforum.org/Projects/s-z/Surgery_Measures_2014/Final_Report.aspx . Accessed February 13, 2017.
11. American Society of Anesthesiologists. RE: CMS-5517-P Medicare Program; Merit-Based Incentive Payment System (MIPS) and Alternative Payment Model (APM) Incentive under the Physician Fee Schedule, and Criteria for Physician-Focused Payment Models, Proposed Rule. Available at:
http://www.asahq.org/~/media/sites/asahq/files/public/quality-and-practice-mgt/2016-06-27-asa-comments-cms-5517p-mips-and-apm.pdf?la=en . Accessed February 13, 2017.
12. Department of Health and Human Services Centers for Medicare and Medicaid Services. 42 CFR Parts 414 and 495 [CMS-5517-FC] Medicare Program; Merit-Based Incentive Payment System (MIPS) and Alternative Payment Model (APM) Incentive Under the Physician Fee Schedule, and Criteria for Physician-Focused Payment Models. Available at:
https://www.gpo.gov/fdsys/pkg/FR-2016-11-04/pdf/2016–25240.pdf . Accessed February 17, 2017.
13. H.R.2—Medicare Access and CHIP Reauthorization Act of 2015. Available at:
https://www.congress.gov/bill/114th-congress/house-bill/2/text . Accessed February 13, 2017.
14. Langham GE, Maheshwari A, Contrera K, You J, Mascha E, Sessler DI. Noninvasive temperature monitoring in postanesthesia care units. Anesthesiology. 2009;111:90–96.
15. Epstein RH, Dexter F, Lopez MG, Ehrenfeld JM. Anesthesiologist staffing considerations consequent to the temporal distribution of hypoxemic episodes in the postanesthesia care unit. Anesth Analg. 2014;119:1322–1333.
16. Goodman D, Ogrinc G, Davies L, et al. Explanation and elaboration of the SQUIRE (Standards for Quality Improvement Reporting Excellence) Guidelines, V.2.0: examples of SQUIRE elements in the healthcare improvement literature. BMJ Qual Saf.2016;25:e7.
17. Sun Z, Honar H, Sessler DI, et al. Intraoperative core temperature patterns, transfusion requirement, and hospital duration in patients warmed with forced air. Anesthesiology. 2015;122:276–285.
18. Trentman TL, Mueller JT, Ruskin KJ, Noble BN, Doyle CA. Adoption of anesthesia information management systems by US anesthesiologists. J Clin Monit Comput. 2011;25:129–135.
19. Centers for Medicare and Medicaid Services. Benchmarks for Measures Included in the Performance Year 2015 Quality and Resource Use Reports.
https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeedbackProgram/Downloads/PY2015-Prior-Year-Benchmarks.pdf (page 46). Accessed February 13, 2017.
20. National Quality Forum. #2681 Perioperative Temperature Management.
http://www.qualityforum.org/ProjectTemplateDownload.aspx?SubmissionID=2681 (see p6 in the document 2681.pdf from the ZIP archive downloaded from the supplied URL). Accessed February 13, 2017.
21. Steelman VM, Perkhounkova YS, Lemke JH. The gap between compliance with the quality performance measure “perioperative temperature management” and normothermia. J Healthc Qual. 2015;37:333–341.
22. Dexter F, Epstein RH. Increased mean time from end of surgery to operating room exit in a historical cohort of cases with prolonged time to extubation. Anesth Analg. 2013;117:1453–1459.
23. Epstein RH, Dexter F, Lopez MG, Ehrenfeld JM. Anesthesiologist staffing considerations consequent to the temporal distribution of hypoxemic episodes in the postanesthesia care unit. Anesth Analg. 2014;119:1322–1333.
24. Stol IS, Ehrenfeld JM, Epstein RH. Technology diffusion of anesthesia information management systems into academic anesthesia departments in the United States. Anesth Analg. 2014;118:644–650.
25. Sandberg WS, Sandberg EH, Seim AR, et al. Real-time checking of electronic anesthesia records for documentation errors and automatically text messaging clinicians improves quality of documentation. Anesth Analg. 2008;106:192–201.
26. Edwards KE, Hagen SM, Hannam J, Kruger C, Yu R, Merry AF. A randomized comparison between records made with an anesthesia information management system and by hand, and evaluation of the Hawthorne effect. Can J Anaesth. 2013;60:990–997.
27. Avidan A, Dotan K, Weissman C, Cohen MJ, Levin PD. Accuracy of manual entry of drug administration data into an anesthesia information management system. Can J Anaesth. 2014;61:979–985.
28. Matsukawa T, Sessler DI, Ozaki M, Hanagata K, Iwashita H, Kumazawa T. Comparison of distal oesophageal temperature with “deep” and tracheal temperatures. Can J Anaesth. 1997;44:433–438.
29. Wachtel RE, Dexter F, Epstein RH, Ledolter J. Meta-analysis of desflurane and propofol average times and variability in times to extubation and following commands. Can J Anaesth. 2011;58:714–724.
30. Dexter F, Bayman EO, Epstein RH. Statistical modeling of average and variability of time to extubation for meta-analysis comparing desflurane to sevoflurane. Anesth Analg. 2010;110:570–580.
31. Dexter F, Epstein RH. Increased mean time from end of surgery to operating room exit in a historical cohort of cases with prolonged time to extubation. Anesth Analg. 2013;117:1453–1459.
32. Larach MG, Brandom BW, Allen GC, Gronert GA, Lehman EB. Malignant hyperthermia deaths related to inadequate temperature monitoring, 2007-2012: a report from the North American malignant hyperthermia registry of the malignant hyperthermia association of the United States. Anesth Analg. 2014;119:1359–1366.
33. Shafer SL, Dexter F, Brull SJ. Deadly heat: economics of continuous temperature monitoring during general anesthesia. Anesth Analg. 2014;119:1235–1237.
34. Madrid E, Urrútia G, Roqué i Figuls M, et al. Active body surface warming systems for preventing complications caused by inadvertent perioperative hypothermia in adults. Cochrane Database Syst Rev. 2016;4:CD009016.
