Abstract
Background/Objectives:
Advanced age is associated with increased susceptibility to acute adverse effects of propofol. The present study aimed to describe patterns of propofol dosing for induction of general anesthesia prior to endotracheal intubation in a nationwide sample of older adults presenting for surgery.
Design:
Retrospective observational study using the Multicenter Perioperative Outcomes Group dataset.
Setting:
36 institutions across the United States.
Participants:
350,766 patients over age 65 who received propofol for general anesthetic induction and endotracheal intubation between 2014 and 2018.
Intervention:
None
Measurements:
Total induction bolus dose of propofol administered.
Results:
The mean (SD) weight-adjusted propofol dose was 1.7 (0.6) mg/kg. The mean prevalent propofol induction dose exceeded the upper bound of what has been described as the typical geriatric dose requirement across every age category examined. The percent of patients receiving propofol induction doses above the described typical geriatric range was 64.8% (95% CI 64.6-65.0), varying from 73.8% among patients age 65-69 to 45.8% among patients age 80 or over.
Conclusion.
The present study of a large multicenter cohort demonstrates that prevalent propofol dosing commonly falls above the published typically required dose range for patients age ≥ 65 in nationwide anesthetic practice. Widespread variability in induction dose administration remains incompletely explained by known patient variables. The nature and clinical consequences of these unexplained dosing decisions remain important topics for further study. Observed discordance between expected and actual induction dosing raises the question of whether there should be reconsideration of widespread provider practice, or alternatively, whether what is published as the typical propofol induction dose range should be revisited.
Keywords: Propofol, Anesthetic Induction, Geriatric Dosing
Introduction:
The US population is aging, with over 16.5% of Americans now over age 65.(1) Older patients account for approximately 40% of the surgical procedures performed each year.(2) Older patients undergoing surgical procedures often have age-related medical conditions and frailty(3-6) which increase risks of adverse outcomes.(7, 8) In addition to the greater medical morbidity that accompanies aging,(9, 10) physiologic changes in metabolism and end-organ function that accompany aging increase the effective potency of anesthetic agents. Propofol sensitivity is increased by approximately 30% to 50% in the elderly compared with younger patients.(11) Even among relatively healthy elderly patients treated under controlled, experimental conditions, advanced age is associated with substantially increased susceptibility to acute adverse effects of propofol.(12)
In recognition of the adverse consequences of advanced age on anesthetic potency, the FDA-approved package insert for propofol recommends adjusting the general anesthesia induction dose among patients > 55 years of age to 1-1.5mg/kg as opposed to the 2-2.5mg/kg recommendation for patients under 55 years of age. Consistent with this recommendation, among generally healthy 60-75 years of age who presented for elective outpatient surgery, EEG-guided general anesthetic states sufficient for endotracheal intubation were reliably induced following a mean propofol dose of 1.27mg/kg.(13) Despite product label guidance and empirical evidence regarding typical dose requirements for geriatric populations, small, single center cohort studies (14-16) suggest that prevalent dosing frequently exceeds the expected typical dose range. Other studies have documented elevated dosing of inhalation agents for older adults during the maintenance period of anesthetics relative to what literature would seem to recommend.(17, 18) Beyond these single-center reports, large, population-based studies of prevalent propofol induction dosing practices for geriatric surgical patients have not been described.
It is well-recognized that doses required by individual patients, or even by the same patient, may vary based on a variety of context-specific factors. Thus, an analysis of actual propofol induction dosing practice among older adults - and how such practice relates to what is deemed by approved labelling to be the typically required range - may generate questions regarding whether current labelling or current practice patterns may be in need of further investigation.
Using a sample of adults ≥ age 65 derived from the Multicenter Perioperative Outcomes Group (MPOG), we describe population-based patterns and predictors of propofol induction dosing practices for geriatric surgical patients. MPOG is a consortium of over 50 academic and community hospitals dedicated to sharing data for quality improvement and research in anesthesiology.(19-21)
Our objective was to describe prevalent propofol induction dosing patterns. Our primary inferential hypothesis was that prevalent Propofol induction dosing among patients ≥ age 65 would exceed the described typical dose range in more than 30% of cases, consistent with prior observational data. Specific analyses sought to a) delineate patterns of weight-adjusted propofol induction dose by age and severity of illness; b) examine predictors of and variation in prevalent propofol dose across patient, institutional, and surgical factors; and c) examine the extent of discontinuous propofol induction dosing at 50 mg increments.
Methods:
This was a retrospective observational study conducted under an approved IRB protocol in collaboration with the Multicenter Perioperative Outcomes Group (MPOG).(22, 23) MPOG is a consortium of over 50 hospitals that maintains a dataset of electronic health record and administrative data from contributing sites across twenty-one states and two countries. The MPOG data collection methods have been previously described (see www.mpog.org).(19, 20) Briefly, MPOG contributing sites conduct automated extraction of perioperative data including patient and procedural characteristics, anesthetic medications, physiologic parameters, and key surgical events.
Contributing sites conduct monthly case-by-case validation of a random sample of submitted data by subject-matter experts. The MPOG Perioperative Clinical Research Committee approved the use of data for this project, and the analytic plan for the present analyses was publicly registered on the website of the Open Science Framework with modifications as noted in the manuscript. The study was conducted in accordance with the Reporting of studies Conducted using Observational Routinely – collected health Data (RECORD) statement, an extension of the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines.(24, 25)
Surgical cases of patients ≥65 years of age who underwent induction of general anesthesia including propofol bolus dosing prior to endotracheal intubation between January 2014 and December 2018 were identified. Only data from centers with consistent data uploads and case validation were included. For patients who underwent multiple anesthetics in the study period, only the index case was included.
Patients were excluded if they were designated as American Society of Anesthesiologists Physical Status (ASA PS) 5 or 6,(26, 27) indicating a moribund or brain-dead patient, respectively. Cases were excluded if the patient a) arrived to the operating room already intubated or b) presented with hemodynamic derangements including severe hypertension or hypotension. Hypertensive patients were excluded for Mean Arterial Pressure (MAP) > 135mmHg or Diastolic Blood Pressure >110mmHg. Hypotensive patients were excluded for MAP < 55mmHg or Systolic Blood Pressure < 90mmHg. Patients receiving anesthetics for minor procedures as determined by the base unit value of anesthesia billing less than or equal to 3 were excluded. Patients who did not undergo endotracheal intubation were excluded. Finally, patients with missing or outlier propofol induction dosing or with other missing or outlier variables were excluded by investigator consensus.
For the selected sample, extracted data were comprised of demographic variables (age, sex, race/ethnicity, anonymized institution ID, and year of surgery); and patient-level clinical characteristics (weight, height, Body-Mass Index categorized into World Health Organization obesity classifications,(28) ASA Physical Status, preoperative mean arterial pressure, the presence of Coronary Artery Disease, and Elixhauser comorbidity status(29)). Data sources for patient and surgical characteristics relied on MPOG-validated, publicly available perioperative electronic health record (EHR) phenotype algorithms.(30) Elixhauser comorbidities were combined into a single comorbidity index based on the van Walraven modification of the Elixhauser classifications.(31) The van Walraven-Elixhauser comorbidity score is a weighted index created from among 30 comorbid conditions that has been demonstrated to independently predict hospital mortality with superior discrimination in comparison to the Charlson Comorbidity Index.(32) Administrative codes were also queried to calculate an index of the likelihood of frailty.(4)
Where laboratory data were available, preoperative albumin level, preoperative hemoglobin levels, and estimated Glomerular Filtration Rate (eGFR) were included. eGFR was calculated using CKD-Epi(33) or using the Cockcroft-Gault formula(34) if race was missing.(35) The eGFR was coded into Chronic Kidney Disease stages in accordance with prior literature.(36)
Surgeries were classified by the specialty of the primary proceduralist and the presence or absence of an emergency modifier in the ASA Physical Status classification. Institutions were classified as academic vs. community based on the presence of an affiliation with a medical school. Classifications of providers within institutions (e.g. solo anesthesiologist, supervised nurse, or supervised resident) were highly imbalanced across institutions and thus were not considered for analysis.
Intraoperative records were queried to calculate the total bolus intravenous anesthetic doses delivered within 15 minutes of induction but ending at surgical incision. In addition to propofol, intravenous anesthetics recorded for the analysis included midazolam, fentanyl, etomidate, ketamine, methohexital, and thiopental. If neuraxial anesthetics (i.e. epidural or spinal anesthetics) were administered concomitantly, this was recorded as a categorical variable. As the present manuscript was aimed at delineating contemporary propofol induction dosing practices, post-induction intraoperative data were not examined.
Statistical Analysis:
Following delineation of the cohort, descriptive statistics (means, standard deviations, medians, interquartile ranges, and frequencies) were computed as relevant.
Cleaning of data:
In addition to contributing-site data validation, study-specific data were examined for outlier values which were considered in relation to clinical plausibility with cut-points derived by agreement among investigators. Missingness was examined for each variable. In accordance with the a priori plan, albumin level demonstrated greater than 40% missingness and was removed from consideration.
Propofol induction dose by age and ASA Physical Status
After initial assessment of data quality and distributions, propofol induction dosing was summarized in relation to what the FDA-approved package insert describes as the typical dosing range across 4 age subgroups (65-69 years, 70-74 years, 75-79 years, and ≥80 years), and within 4 categories of comorbidity status (ASA Physical Status 1, 2, 3, and 4) for a total of 16 age by severity of illness sub-categories. Within each sub-category, mean, SD of propofol dose, and percent dose in excess of the described typical dosing range (i.e. >1.5mg/kg) were calculated.
Independent Predictors of Propofol induction dose
To determine predictors of propofol induction dose and derive the adjusted rates of doses in excess of the package insert’s listed typical dose requirement, we fit suitable hierarchical logistic regression models that accounted for the clustering of the data within institutions to: 1) identify covariates (fixed effects) including patient, institutional, and surgical characteristics that were significantly associated with excess prevalent dose; 2) predict the random effects of institutions after adjusting for significant covariates; and 3) estimate the standardized institution rates of prevalent dosing in excess of the typically described requirement for institution-level comparisons.
In response to reviewer requests in the context of the peer review process, we further conducted a hierarchical linear regression model in which the outcome of propofol dose was maintained as a continuous variable without reference to the FDA-approved package insert so as to understand factors associated with actual propofol induction dose across the range of prevalent dosing behaviors and in units of total milligrams to enhance clinical interpretability for practitioners.
In these multivariable models, we evaluated effect modification by age sub-groups and ASA Physical Status to investigate whether combinations of age and comorbidity affected providers’ choice of propofol induction dose. Before regression models were constructed, the continuous explanatory variables under consideration for model inclusion were assessed for collinearity using the variance inflation factor with a threshold of 10 suggestive of substantial collinearity. All analyses were performed using the SAS statistical software, version 9.4 (Cary, NC). Within SAS, we fitted hierarchical models using the PROC GLIMMIX procedure. All statistical tests’ p-values and confidence intervals were two-sided with a significance level for the primary outcome of 0.05.
Institution specific Propofol dosing Rates in excess of FDA’s typical range:
For institution effects, we compared predicted non-adherence rates (i.e. institution-specific or predicted) using both random institution effect and the fixed effect regression estimates to the fixed effects only (population-average or expected) rates. The intra-institution comparison of the two rates was used as a measure of institution performance in rates of incremental dosing in excess of the published typical range.(37, 38)
Discontinuous Propofol Dosing at 50mg Increments
We examined non-weight-adjusted anchoring effects in propofol dosing by measuring the tendency for induction doses to cluster at 50 mg increments (i.e. 50, 100, 150, 200, 250, and 300 mg). The percent of total doses occurring at each of these 50mg increments was reported.
Sample size considerations:
For the primary analysis estimating the prevalence of dosing in excess of the described typical dose range, a hypothetical subset of 10,000 cases with actual incremental dosing rates of 30% would produce a two-sided 95% CI of 29.1-30.9%. For such a hypothetical subset, power would exceed 99% to detect a hypothetical difference of 2% between observed dose-range-adherence and the 30% a priori benchmark, assuming an alpha of 0.05.
Results:
Over 2.1 million patients aged ≥65 underwent anesthetics during the observation period. After exclusions, a final sample of 350,766 patients from 36 institutions were included in the present analyses (see consort diagram Figure 1). The mean (SD) age was 73 (6.4), with 16.8% of the cohort over age 80, 49.6% female, and 81.3% identified as white. Ethnic identification was poorly penetrant, with only 1.4% of the cohort identified as Hispanic. Characteristics of the included cohort are provided in Table 1. Due to non-random and significant missingness of preoperative laboratory data, only patients with complete data regarding hemoglobin and estimated GFR were included in multivariable models.
Figure 1:
Consort Diagram of Inclusion and Exclusion Criteria
Table 1.
Summary statistics of variables by propofol dose status
| Propofol Dose Above Range (>1.5 mg/kg) | |||
|---|---|---|---|
| No (N = 123492) |
Yes (N = 227274) |
Total (N = 350766) |
|
| Age | |||
| 65 - 69 | 34081 (27.60%) | 96336 (42.39%) | 130417 (37.18%) |
| 70 - 74 | 31762 (25.72%) | 65298 (28.73%) | 97060 (27.67%) |
| 75 - 79 | 25821 (20.91%) | 38742 (17.05%) | 64563 (18.41%) |
| 80 or Over | 31828 (25.77%) | 26898 (11.84%) | 58726 (16.74%) |
| Sex | |||
| Female | 57306 (46.40%) | 116533 (51.27%) | 173839 (49.56%) |
| Male | 66186 (53.60%) | 110741 (48.73%) | 176927 (50.44%) |
| Race | |||
| Black | 8710 ( 7.05%) | 15288 ( 6.73%) | 23998 ( 6.84%) |
| Unknown/other Race | 14328 (11.60%) | 27180 (11.96%) | 41508 (11.83%) |
| White | 100454 (81.34%) | 184806 (81.31%) | 285260 (81.32%) |
| Ethnicity | |||
| Hispanic Ethnicity | 1464 ( 1.19%) | 3368 ( 1.48%) | 4832 ( 1.38%) |
| Non-Hispanic Ethnicity | 113766 (92.12%) | 208745 (91.85%) | 322511 (91.94%) |
| Unknown Ethnicity | 8262 ( 6.69%) | 15161 ( 6.67%) | 23423 ( 6.68%) |
| Year of Surgery | |||
| 2014 | 16630 (13.47%) | 27978 (12.31%) | 44608 (12.72%) |
| 2015 | 22328 (18.08%) | 37942 (16.69%) | 60270 (17.18%) |
| 2016 | 24466 (19.81%) | 43674 (19.22%) | 68140 (19.43%) |
| 2017 | 29356 (23.77%) | 57333 (25.23%) | 86689 (24.71%) |
| 2018 | 30712 (24.87%) | 60347 (26.55%) | 91059 (25.96%) |
| Weight(kg) | |||
| N (N Missing) | 123492 (0) | 227274 (0) | 350766 (0) |
| Mean (SD) | 85.67 (19.14) | 78.41 (17.33) | 80.97 (18.32) |
| Median (IQR) | 83.9 (71.7 – 99.6) | 77.1 (65.3 – 89.8) | 79.4 (67.6 – 92.6) |
| Height cm | |||
| N (N Missing) | 118147 (5345) | 218237 (9037) | 336384 (14382) |
| Mean (SD) | 169.47 (10.67) | 168.22 (10.48) | 168.66 (10.56) |
| Median (IQR) | 170.2 (161.3 – 177.8) | 167.6 (160.0 – 175.6) | 167.9 (160.0 – 177.8) |
| WHO Obesity Class based on BMI | |||
| Missing | 5237 | 8864 | 14101 |
| Class I obesity (BMI 30 to <35) | 29331 (24.80%) | 41793 (19.14%) | 71124 (21.13%) |
| Class II obesity (BMI 35 to <40) | 14546 (12.30%) | 14963 ( 6.85%) | 29509 ( 8.77%) |
| Class III obesity (BMI ≥40) | 7982 ( 6.75%) | 5501 ( 2.52%) | 13483 ( 4.00%) |
| Normal weight (BMI 18.5 to <25) | 25694 (21.73%) | 71786 (32.87%) | 97480 (28.95%) |
| Overweight (BMI (25 to <30) | 40181 (33.98%) | 82611 (37.82%) | 122792 (36.47%) |
| Underweight (BMI <18.5) | 521 ( 0.44%) | 1756 ( 0.80%) | 2277 ( 0.68%) |
| ASA Physical Status | |||
| Missing | 2 | 3 | 5 |
| 1 | 379 ( 0.31%) | 2042 ( 0.90%) | 2421 ( 0.69%) |
| 2 | 21745 (17.61%) | 77474 (34.09%) | 99219 (28.29%) |
| 3 | 79122 (64.07%) | 135404 (59.58%) | 214526 (61.16%) |
| 4 | 22244 (18.01%) | 12351 ( 5.43%) | 34595 ( 9.86%) |
| Emergency Modifier | |||
| Missing | 2 | 2 | 4 |
| Emergency | 6978 ( 5.65%) | 7456 ( 3.28%) | 14434 ( 4.12%) |
| Non-Emergency | 116512 (94.35%) | 219816 (96.72%) | 336328 (95.88%) |
| Neuraxial Anesthetic | |||
| No | 122233 (98.98%) | 224128 (98.62%) | 346361 (98.74%) |
| Yes | 1259 ( 1.02%) | 3146 ( 1.38%) | 4405 ( 1.26%) |
| van Walraven Comorbidity Index | |||
| N (N Missing) | 121523 (1969) | 224680 (2594) | 346203 (4563) |
| Mean (SD) | 7.94 (9.83) | 5.85 (8.54) | 6.59 (9.07) |
| Median (IQR) | 5.0 (0.0 – 14.0) | 3.0 (0.0 – 10.0) | 4.0 (0.0 – 11.0) |
| Frailty Index | |||
| Missing | 1969 | 2594 | 4563 |
| High | 1360 (1.12%) | 1346 (0.60%) | 2706 (0.78%) |
| Intermediate | 12863 (10.58%) | 21186 (9.43%) | 34049 (9.83%) |
| Low | 107300 (88.30%) | 202148 (89.97%) | 309448 (89.38%) |
| Coronary Artery Disease | |||
| Missing | 1969 | 2594 | 4563 |
| No | 86008 (70.78%) | 188756 (84.01%) | 274764 (79.36%) |
| Yes | 35515 (29.22%) | 35924 (15.99%) | 71439 (20.64%) |
| Chronic Kidney Disease | |||
| Missing | 33033 | 67618 | 100651 |
| Normal eGFR | 11467 (12.68%) | 28085 (17.59%) | 39552 (15.81%) |
| Stage 2 | 46662 (51.58%) | 93036 (58.27%) | 139698 (55.85%) |
| Stage 3a | 18652 (20.62%) | 25750 (16.13%) | 44402 (17.75%) |
| Stage 3b | 9991 (11.04%) | 9968 ( 6.24%) | 19959 ( 7.98%) |
| Stage 4 | 3687 ( 4.08%) | 2817 ( 1.76%) | 6504 ( 2.60%) |
| Anemia | |||
| Missing | 25877 | 55638 | 81515 |
| Mild | 34834 (35.69%) | 52325 (30.49%) | 87159 (32.37%) |
| Moderate to Severe | 16739 (17.15%) | 16740 (9.75%) | 33479 (12.43%) |
| Normal | 46042 (47.17%) | 102571 (59.76%) | 148613 (55.19%) |
| Baseline Mean Arterial Pressure | |||
| Mean (SD) | 97.42 (14.83) | 99.71 (14.42) | 98.91 (14.61) |
| Median (IQR) | 97.0 (87.0 – 107.0) | 99.0 (90.0 – 109.0) | 99.0 (89.0 – 109.0) |
| Duration of Surgery in Minutes | |||
| N (N Missing) | 123230 (262) | 226980 (294) | 350210 (556) |
| Mean (SD) | 216.24 (130.08) | 206.89 (118.26) | 210.18 (122.63) |
| Median (IQR) | 188.0 (123.0 – 283.0) | 180.0 (123.0 – 262.0) | 183.0 (123.0 – 269.0) |
| Anesthesia Base Units | |||
| Minor (≤5 Base Units) | 29918 (24.23%) | 56809 (25.00%) | 86727 (24.73%) |
| Major (>5 Base Units) | 93574 (75.77%) | 170465 (75.00%) | 264039 (75.27%) |
| Institution Type | |||
| Academic Hospital | 101706 (82.36%) | 192473 (84.69%) | 294179 (83.87%) |
| Community Hospital | 21786 (17.64%) | 34801 (15.31%) | 56587 (16.13%) |
| Surgical Service of Proceduralist | |||
| Cardiothoracic | 11956 ( 9.68%) | 7337 ( 3.23%) | 19293 ( 5.50%) |
| General Surgery | 17561 (14.22%) | 44462 (19.56%) | 62023 (17.68%) |
| Medical - Gastroenterology | 7330 ( 5.94%) | 6023 ( 2.65%) | 13353 ( 3.81%) |
| Medical - Other Subspecialty | 2821 ( 2.28%) | 3938 ( 1.73%) | 006759 (01.93%) |
| Neurosurgery | 10454 ( 8.47%) | 22486 ( 9.89%) | 32940 ( 9.39%) |
| Gynecology | 3474 ( 2.81%) | 11793 ( 5.19%) | 15267 ( 4.35%) |
| Ophthalmology | 1585 ( 1.28%) | 2423 ( 1.07%) | 4008 ( 1.14%) |
| Oral/Maxillofacial/Otolaryngology | 8120 ( 6.58%) | 20702 ( 9.11%) | 28822 ( 8.22%) |
| Orthopedics | 19635 (15.90%) | 34777 (15.30%) | 54412 (15.51%) |
| Other | 20230 (16.38%) | 30573 (13.45%) | 50803 (14.48%) |
| Plastic Surgery | 1303 ( 1.06%) | 4874 ( 2.14%) | 6177 ( 1.76%) |
| Radiology | 883 ( 0.72%) | 2216 ( 0.98%) | 3099 ( 0.88%) |
| Thoracic | 4462 ( 3.61%) | 9535 ( 4.20%) | 13997 ( 3.99%) |
| Transplant | 666 ( 0.54%) | 1434 ( 0.63%) | 2100 ( 0.60%) |
| Trauma | 901 ( 0.73%) | 1200 ( 0.53%) | 2101 ( 0.60%) |
| Urology | 6153 ( 4.98%) | 18193 ( 8.00%) | 24346 ( 6.94%) |
| Vascular | 5958 ( 4.82%) | 5308 ( 2.34%) | 11266 ( 3.21%) |
| Additional Induction Medications | |||
| Etomidate | |||
| Dose = 0 mg | 119347 (96.64%) | 226922 (99.85%) | 346269 (98.72%) |
| Dose >0 mg | 4145 (3.36%) | 352 (0.15%) | 4497 (1.28%) |
| Ketamine | |||
| Dose = 0 mg | 115981 (93.92%) | 216654 (95.33%) | 332635 (94.83%) |
| Dose >0 mg | 7511 (6.08%) | 10620 (4.67%) | 18131 (5.17%) |
| Fentanyl (mcg) | |||
| Mean (SD) | 89.70 (69.97) | 87.73 (59.96) | 88.42 (63.67) |
| Median (IQR) | 100.0 (50.0 – 100.0) | 100.0 (50.0 – 100.0) | 100.0 (50.0 – 100.0) |
| Midazolam (mg) | |||
| Mean (SD) | 0.68 (1.09) | 0.64 (0.94) | 0.66 (0.99) |
| Median (IQR) | 0.0 (0.0 – 2.0) | 0.0 (0.0 – 2.0) | 0.0 (0.0 – 2.0) |
Propofol induction dose by age and ASA Physical Status
The percent of patients receiving propofol induction doses in excess of the typically described range was 64.8% (95% CI 64.6-65.0), with the mean (SD) weight-adjusted propofol dose being 1.7 (0.6) mg/kg. As shown in Figure 2, mean propofol dose exceeded the expected typical range across every age by ASA PS score classification except for patients with ASA Physical Status 4. Within ASA PS 4, the proportion of patients that received propofol induction doses in excess of the typical range across the four age bins ranged from 43.1% in those age 65-69 years to 26.9% in those age 80 years or older. Propofol induction doses by age and ASA PS Score are summarized in Figure 2 (see also histograms of weight-adjusted dose distributions displayed in Supplemental Figure S1 and summaries as listed in Supplemental Table S2).
Figure 2:
Boxplots of Propofol Dose by Age and ASA Physical Status. Diamonds show the mean and shaded boxes show the 25-75% range with whiskers at 1.5 times the upper and lower bounds. The horizontal dotted line demonstrates the upper limit of the package insert’s listed typical dosing requirement.
Independent predictors of propofol dose:
In multivariable modelling, the prevalence of propofol induction dose in excess of the typically described range increased from 2014 through 2018. Such doses were more likely in males and patients identifying as Hispanic. In prior literature, it has been suggested that black vs. white race may be implicated in differing clinical responses to propofol infusions,(39) but we observed no significant differences in prevalent dosing above the typical range associated with black vs. white race. Increasing age and increasing obesity were negatively associated with exceeding the described range as was the use of adjunctive sedative medications. Doses exceeding the described range were less likely in sicker patients based on ASA Physical Status, coronary artery disease, van Walraven comorbidity index, chronic kidney disease, and lower preoperative hemoglobin levels. While providers thus appeared to decrease induction dose for both older age and higher ASA PS score independently, the multiplicative interaction terms of these factors were not significant except for the interaction of age 80 or older with ASA Physical Status score of 4. The full multivariable results are listed in Tables 2 and Supplemental Table S3.
Table 2:
Multivariable Analysis Evaluating Perioperative Factors Associated with Odds of Propofol Dose Above Range (N=228,492 with preoperative laboratory data).
| Variable | Odds Ratio (95% CI) | p-Value |
|---|---|---|
| Age | <0.0001* | |
| 65 - 69 | Ref | |
| 70 - 74 | 0.603 (0.388 to 0.939) | 0.0253 |
| 75 - 79 | 0.419 (0.229 to 0.766) | 0.0047 |
| 80 or Over | 0.128 (0.065 to 0.253) | <0.0001 |
| Sex (Female vs Male) | 0.744 (0.722 to 0.767) | <0.0001 |
| Race | <0.0001* | |
| Black | 1.036 (0.996 to 1.077) | 0.0788 |
| Unknown/other | 0.772 (0.736 to 0.81) | <0.0001 |
| White | Ref | |
| Ethnicity | <0.0001* | |
| Hispanic | 1.204 (1.106 to 1.31) | <0.0001 |
| Unknown | 1.277 (1.192 to 1.367) | <0.0001 |
| Non-Hispanic | Ref | |
| Year of Surgery | <0.0001* | |
| 2014 | Ref | |
| 2015 | 1.029 (0.992 to 1.068) | 0.1266 |
| 2016 | 1.029 (0.992 to 1.067) | 0.1257 |
| 2017 | 1.166 (1.125 to 1.208) | <0.0001 |
| 2018 | 1.186 (1.145 to 1.228) | <0.0001 |
| Weight, kg | 0.969 (0.967 to 0.971) | <0.0001 |
| Height (cm) a | 1.000 (0.997 to 1.002) | 0.8078 |
| WHO Obesity Class | <0.0001* | |
| Normal weight (BMI 18.5 to <25) | Ref | |
| Overweight (BMI (25 to <30) | 0.916 (0.882 to 0.951) | <0.0001 |
| Class I obesity (BMI 30 to <35) | 0.831 (0.78 to 0.886) | <0.0001 |
| Class II obesity (BMI 35 to <40) | 0.806 (0.734 to 0.884) | <0.0001 |
| Class III obesity (BMI ≥40) | 0.739 (0.651 to 0.839) | <0.0001 |
| Underweight (BMI <18.5) | 1.029 (0.902 to 1.174) | 0.6713 |
| ASA Physical Status | <0.0001* | |
| 1 | Ref | |
| 2 | 0.793 (0.607 to 1.035) | 0.0873 |
| 3 | 0.539 (0.414 to 0.703) | <0.0001 |
| 4 | 0.225 (0.172 to 0.294) | <0.0001 |
| Age x ASA Physical Status Interaction | <0.0001* | |
| 70 – 74 x 2 | 1.075 (0.688 to 1.679) | 0.7504 |
| 70 – 74 x 3 | 1.104 (0.708 to 1.72) | 0.6624 |
| 70 – 74 x 4 | 1.268 (0.809 to 1.987) | 0.3006 |
| 70 – 74 x 1 | Ref | |
| 75 – 79 x 2 | 0.955 (0.52 to 1.751) | 0.8804 |
| 75 – 79 x 3 | 1.072 (0.586 to 1.962) | 0.8208 |
| 75 – 79 x 4 | 1.366 (0.743 to 2.511) | 0.3154 |
| 75 – 79 x 1 | Ref | |
| 80 or Over x 2 | 1.391 (0.705 to 2.744) | 0.3419 |
| 80 or Over x 3 | 1.64 (0.833 to 3.229) | 0.1523 |
| 80 or Over x 4 | 2.463 (1.246 to 4.867) | 0.0095 |
| 80 or Over x 1 | Ref | |
| Emergency Modifier | 0.697 (0.667 to 0.729) | <0.0001 |
| Neuraxial Anesthesia, Yes | 1.371 (1.26 to 1.491) | <0.0001 |
| van Walraven Comorbidity Index | 0.991 (0.99 to 0.992) | <0.0001 |
| Frailty Index | <0.0001* | |
| High | 0.828 (0.749 to 0.915) | 0.0002 |
| Intermediate | 1.04 (1.007 to 1.074) | 0.017 |
| Low | Ref | |
| Coronary Artery Disease | 0.76 (0.741 to 0.778) | <0.0001 |
| Chronic Kidney Disease | <0.0001* | |
| Stage 2 | 1.033 (1.003 to 1.065) | 0.0302 |
| Stage 3a | 0.957 (0.924 to 0.991) | 0.0143 |
| Stage 3b | 0.879 (0.841 to 0.918) | <0.0001 |
| Stage 4 | 0.804 (0.753 to 0.858) | <0.0001 |
| Normal eGFR | Ref | |
| Anemia | <0.0001* | |
| Mild | 0.831 (0.812 to 0.849) | <0.0001 |
| Moderate to Severe | 0.642 (0.621 to 0.663) | <0.0001 |
| Normal | Ref | |
| Baseline Mean Arterial Pressure | 1.012 (1.011 to 1.013) | <0.0001 |
| Duration of Surgery in Minutes b | 0.997 (0.997 to 0.998) | <0.0001 |
| Anesthesia Base Units >5 | 1.1 (1.069 to 1.131) | <0.0001 |
| Academic vs. Community Hospital | 1.064 (1.033 to 1.095) | <0.0001 |
| Surgical Service of Proceduralist | <0.0001* | |
| General Surgery | Ref | |
| Cardiothoracic | 0.344 (0.328 to 0.36) | <0.0001 |
| Medical - gastroenterology | 0.437 (0.411 to 0.463) | <0.0001 |
| Medical Subspecialty | 0.606 (0.563 to 0.652) | <0.0001 |
| Neurosurgery | 0.803 (0.773 to 0.833) | <0.0001 |
| Gynecology | 1.078 (1.02 to 1.138) | 0.0072 |
| Ophthalmology | 0.525 (0.449 to 0.614) | <0.0001 |
| Oral/Maxillofacial, ENT | 0.82 (0.782 to 0.859) | <0.0001 |
| Orthopedics | 0.703 (0.68 to 0.726) | <0.0001 |
| Other | 0.63 (0.606 to 0.655) | <0.0001 |
| Plastic Surgery | 1.138 (1.029 to 1.26) | 0.0122 |
| Radiology | 1.16 (1.052 to 1.279) | 0.0029 |
| Thoracic | 0.755 (0.719 to 0.793) | <0.0001 |
| Transplant | 0.629 (0.535 to 0.739) | <0.0001 |
| Trauma | 0.592 (0.531 to 0.66) | <0.0001 |
| Urology | 1.071 (1.024 to 1.12) | 0.0028 |
| Vascular | 0.462 (0.438 to 0.488) | <0.0001 |
| Etomidate (Yes vs No) | 0.067 (0.059 to 0.077) | <0.0001 |
| Ketamine (Yes vs No) | 0.573 (0.55 to 0.598) | <0.0001 |
| Midazolam (mg) | 0.886 (0.876 to 0.895) | <0.0001 |
| Fentanyl (mcg) c | 0.99 (0.986 to 0.994) | <0.0001 |
Note: Individual Institution was treated as a random effect to account for possible similarity among patients cared at each institute. A total of 228492 patients were included in the above analysis
: Type III test p-value
: per 10 cm increase
per 10 min increase.
: per 25 mcg increase.
Regarding absolute propofol doses, factors associated with a mean decrease of at least 10mg of propofol at induction included being over age 75, ASA Physical Status 4, and presenting for Cardiothoracic, Vascular, or GI procedures.
Of note, while the propofol package insert suggests the slow administration of multiple doses of propofol over the course of a geriatric general anesthetic induction, the large majority (282,502 or 81%) of the 350,766 included cases in our cohort showed recorded induction doses charted as a single bolus dose, a finding that should be interpreted cautiously since closely-timed doses may be recorded as one administration. Patients charted as receiving a single bolus dose vs. multiple doses were similar in respect to age (mean 72.9 vs 73.3 years; median age of both groups = 72 years).
Institution-specific Rates of Dosing in excess of FDA’s typical range:
In unadjusted analysis, significant variations in care patterns were demonstrated, with incremental dose rates ranging from 38% to 82% across centers, a difference that cannot be explained solely by differences in the distributions of ASA Physical Status scores or of age (See Figure 3). After adjusting for case-mix and patient characteristics, we observed 4 institutions with random effects predicted incremental dose rates that differed by more than 1% of their fixed effects or population averaged estimates. For these outlying institutions, the actual versus expected rate of incremental dose rates in excess of FDA’s typical range varied from a high of 106% to a low of 92% of predicted.
Figure 3:
Percent of patients receiving Propofol Incremental Induction Dose Above 1.5 mg/kg by Institution is Shown on the Y-axis at left. Shading shows distribution of ASA Physical Status by Institution. Red Dots Demonstrate Mean Age by Institution on the Y-axis at right.
Propofol Dose Discontinuity at 50mg Increments
Total induction doses of propofol demonstrated anchoring at 50mg increments in 60.9% of cases, with more than 1 in 5 doses equaling 150mg. A histogram of all propofol doses by 10mg increments is included, demonstrating multi-modal anchoring across the breadth of the distribution (see Supplemental Figure S4).
Discussion:
This is the first multicenter study to examine propofol dosing practices in relation to the medication label’s typical dose range for older patients undergoing surgical procedures in the US, and the present analysis confirms our primary hypothesis that there is widespread discordance between current geriatric anesthetic practice in the United States and the package label’s description of the typical dose range. Observed doses exceeded the typical dose range by a mean of 13% overall, and significant numbers of patients received higher than expected induction doses regardless of age and comorbidity. Remarkably, over one quarter of the sickest and oldest patients in our sample (i.e. ASA PS 4 patients ≥ 80 years of age) received propofol exposures at induction that would be considered above the typical expected dose requirement for a cohort of healthy patients 25 years younger. While it remains unknown what threshold of propofol induction dosing outside of typical ranges may lead to untoward downstream effects, the discordance between what the package-insert states to be “typical” and the prevalent anesthetic induction dosing practice that we have documented raises the question as to whether the FDA-approved label or the prevalent practice pattern is in need of revision.
Regarding institution-specific behaviors, while we saw variation across institutions in unadjusted proportions of patients receiving incremental doses beyond the typical range, those differences were less apparent in adjusted analyses that accounted for differences in patient mix and procedure type. Still, institutional differences of greater than 1% observed over expected incremental dosing do persist in some centers and appear to be robust within the limits of our adjusted modeling. These results should be interpreted cautiously as they were not the primary inferential hypothesis of the study and should be understood as hypothesis generating rather than definitive evidence of care disparities across institutions.
Regarding dose anchoring, our descriptive analysis demonstrates significant discontinuities in the choice of administered induction dose at intervals of 50mg. More than half of all patients received either 100mg, 150mg, or 200mg of propofol. We conjecture that such anchoring is likely to exacerbate connections between patient weight and rate of dosing in excess of the typical range. For example, a 150mg induction dose discontinuously becomes an incremental dose above typical range for patients weighing less than 100kg but remains concordant within the FDA described range for heavier patients. It may thus prove important to account for this multi-modal, irregular distribution in the design of future outcomes studies that seek to elucidate effects of propofol dosing patterns on important patient outcomes.
Limitations:
Several limitations of the present analysis should be noted. First, our conclusions are based on documented medication dosing that may contain inaccuracies in relation to actual practice. In our cohort, a minority (19%) of cases were noted to have more than one bolus induction dose of propofol, but some induction doses that were slowly titrated may have been added together during the process of clinical documentation. While it is certainly the case that some providers engage in careful titration of induction doses, based on our present data and our own clinical experience, such titration is far from routine. While similar observational studies are subject to recording accuracy, MPOG data collection standards include a robust multicenter validation schema, incorporating case-by-case monthly validations at each participating center, and these data have been used in a variety of high quality observational studies.(40, 41)
Another possible limitation of our analysis may stem from our selection of four coarsened categories for age in our assessments of incremental dosing prevalence. While it is possible there could be a loss of information by grouping ages into sub-categories, categorizing age should allow for detecting trends, can be useful for identifying thresholds across age, and allows one to model non-linearities between age and incremental dose. While we were cognizant of the hazards of binning ages into sub-categories, we chose this method of displaying data so as to improve interpretability of the data as they relate to age and comorbid status and thereby to promote a better understanding of the interactions among these two factors.
It is also important to note that the typical dose range, as described on the package label, does not include a description of the shape of the typical dose range curve. Nevertheless, regardless of the distribution’s shape, it remains inconsistent with what our analysis shows to be the prevalent induction dosing used in clinical practice. Well less than 50% of geriatric patients are receiving an induction dose in the listed “typical” range, and this observation demonstrates a remarkable discordance that remains to be understood in future work.
Another issue to be highlighted is that while we included non-propofol sedatives in our modeling, it is not clear how to interpret their concomitant dosing in relation to an expected propofol dose range. Patients receiving concomitant sedative medications were less likely to receive propofol doses above the listed typically expected range, suggesting that practitioners anticipate desired clinical endpoints to drive dose adjustment.
Finally, the clinical relevance of our observation regarding induction dosing for patient-centered outcomes remains unclear. It would be overly simplistic to extrapolate that older patients are being “overdosed.” Future studies of appropriately matched patients with outcomes data are required to understand the potential for and extent of any differences in relevant clinical course that may be attributable to the propofol induction dosage patterns described here. The effort to define clinical phenotypes that may define varying degrees of patient resilience to anesthetic induction dosing practices remains an area of great importance, with opportunities for future collaboration among geriatricians and anesthesiologists.
In conclusion, the present study provides the broadest and most thorough documentation to date of the widespread use of propofol in excess of the published typical dose range during the conduct of modern general anesthetic induction of elderly surgical patients. The potential significance of this discordance on patient outcomes - both in general and within particularly vulnerable subsets of the elderly – deserves dedicated study. At present, it remains unclear whether a change in drug labels or a change in prevalent anesthetic induction dosing practice may be warranted.
Supplementary Material
Supplemental Figure S1: Histograms of Propofol Induction Dose by Age-Bin and ASA Physical Status. Vertical dotted line demonstrates the upper limit of the listed typical dosing range.
Supplemental Table S2: Table of Absolute and Weight Adjusted Propofol Induction Dose by Age and ASA Physical Status.
Supplemental Table S3: Table: Multivariable Analysis Evaluating Perioperative Factors Associated with Variability in Absolute Propofol Dose Received at Induction.
Supplemental Figure S4: Histogram of Absolute Propofol Dose in Milligrams for the Entire Dataset.
KEY POINTS:
In a nationwide sample of propofol induction dosing practices for patients over age 65 undergoing general anesthesia, almost two-thirds of administered doses were discordant with published typical dose requirements for this age group.
Variation in prevalent propofol induction dosing across institutions is not explained solely by differences in patient populations.
The range of propofol induction doses commonly used in modern anesthetic practice is broad and is not fully explained by demographic and comorbidity variables available in the present dataset. Future investigations should endeavor to understand the nature of these highly variable dosing decisions and how such dosing may relate to important clinical outcomes.
WHY DOES THIS PAPER MATTER:
This multicenter investigation demonstrates discordance between clinical practices and the expected typical dose range for propofol induction in older adults. It is a key step in the ongoing effort to identify opportunities to improve anesthetic care of older adults.
Acknowledgements:
The authors wish to acknowledge the invaluable contributions of Michelle Romanowski for her technical expertise in developing and modifying database queries.
The authors also acknowledge the valuable contribution to protocol development, data acquisition and final article review by Multicenter Perioperative Outcomes Group Perioperative Clinical Research Committee collaborators, including:
Michael F. Aziz, MD; Department of Anesthesiology & Perioperative Medicine, Oregon Health & Science University; Portland, Oregon
Robert M. Craft, MD; Department of Anesthesiology, University of Tennessee Medical Center; Knoxville, Tennessee
Adit A. Ginde, MD MPH; Department of Anesthesiology, University of Colorado; Aurora, Colorado
William Hightower, MD; Department of Anesthesiology, Pain Management and Perioperative Medicine; Henry Ford Health System; West Bloomfield Township, Michigan
Bhiken I. Naik, MBBCh; Departments of Anesthesiology and Neurological Surgery, University of Virginia; Charlottesville, Virginia
Zachary A. Turnbull, MD; Department of Anesthesiology, New York Presbyterian–Weill Cornell; New York, New York
Wilton A. van Klei, MD PhD; Department of Anesthesiology, University Medical Center Utrecht; Utrecht, Netherlands
Funding:
This work was supported in part by grant R01AG059607 from the National Institute on Aging and CTSA Grant UL1 RR024139 from the National Center for Advancing Translational Sciences (NCATS). The content is solely the responsibility of the authors and does not necessarily represent the policy or views of the NIH, NIA, NCATS, or the United States Government.
Support for underlying electronic health record data collection was provided, in part, by Blue Cross and Blue Shield of Michigan (BCBSM) and Blue Care Network as part of the BCBSM Value Partnerships program for contributing hospitals in the State of Michigan. Although Blue Cross Blue Shield of Michigan and the Multicenter Perioperative Outcomes Group work collaboratively, the opinions, beliefs, and viewpoints expressed by the authors do not necessarily reflect the opinions, beliefs, and viewpoints of BCBSM or any of its employees.
Dr. Schonberger reports owning stock in Johnson & Johnson. Dr. Schonberger reports that his institution receives research funding from Merck, Inc. on a project with which he is associated.
Dr. Shah reports that his institution receives grants from Blue Cross Blue Shield Michigan, grants from Apple Inc, grants from Edwards Lifesciences, grants from Merck Inc, and grants from the American Heart Association for work with which he is associated.
Dr. Mathis reports grants from National Institute of Health - NHLBI (K01-HL141701), outside the submitted work.
Footnotes
Sponsors’ Role: The sponsors had no role in the design, methods, subject recruitment, data collections, analysis and preparation of the paper.
REFERENCES:
- 1.United States Census Bureau; Quick Facts - Persons 65 years and over, percent; url: https://www.census.gov/quickfacts/fact/table/US/AGE775219 as accessed June 25, 2020.
- 2.Popovic JR. 1999 National Hospital Discharge Survey: annual summary with detailed diagnosis and procedure data. Vital Health Stat 13. 2001;2001September;(151):i–v. [DOI] [PubMed] [Google Scholar]
- 3.Chen X, Mao G, Leng SX. Frailty syndrome: an overview. Clin Interv Aging. 2014;9:433–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gilbert T, Neuburger J, Kraindler J, Keeble E, Smith P, Ariti C, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. The Lancet. 2018;391(10132):1775–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lin H-S, Watts JN, Peel NM, Hubbard RE. Frailty and post-operative outcomes in older surgical patients: a systematic review. BMC Geriatrics. 2016;16(1):157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McIsaac DI, Moloo H, Bryson GL, van Walraven C. The Association of Frailty With Outcomes and Resource Use After Emergency General Surgery: A Population-Based Cohort Study. Anesthesia & Analgesia. 2017;124(5):1653–61. [DOI] [PubMed] [Google Scholar]
- 7.Dodson JA, Chaudhry SI, Krumholz HM. Time for a New Approach to Studying Older People with Ischemic Heart Disease. J Am Geriatr Soc. 2017;65(11):2349–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Turrentine F, Wang H, Simpson V, Jones R. Surgical Risk Factors, Morbidity, and Mortality in Elderly Patients. Journal of the American College of Surgeons. 2006;203(6):865–77. [DOI] [PubMed] [Google Scholar]
- 9.Deiner S, Westlake B, Dutton RP. Patterns of Surgical Care and Complications in Elderly Adults. J Am Geriatr Soc. 2014;14(10):12794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Polanczyk CA, Marcantonio E, Goldman L, Rohde LE, Orav J, Mangione CM, et al. Impact of age on perioperative complications and length of stay in patients undergoing noncardiac surgery. Ann Intern Med. 2001;134(11304103):637–43. [DOI] [PubMed] [Google Scholar]
- 11.Schnider TW, Minto CF, Shafer SL, Gambus PL, Andresen C, Goodale DB, et al. The influence of age on propofol pharmacodynamics. Anesthesiology. 1999;90(6):1502–16. [DOI] [PubMed] [Google Scholar]
- 12.Dundee JW, Robinson FP, McCollum JSC, Patterson CC. Sensitivity to propofol in the elderly. Anaesthesia. 1986;41(5):482–5. [DOI] [PubMed] [Google Scholar]
- 13.Riad W, Schreiber M, Saeed AB. Monitoring with EEG entropy decreases propofol requirement and maintains cardiovascular stability during induction of anaesthesia in elderly patients. European Journal of Anaesthesiology;24(8):684–8. [DOI] [PubMed] [Google Scholar]
- 14.Akhtar S, Liu J, Heng J, Dai F, Schonberger RB, Burg MM. Does intravenous induction dosing among patients undergoing gastrointestinal surgical procedures follow current recommendations: a study of contemporary practice. J Clin Anesth. 2016;33:208–15. [DOI] [PubMed] [Google Scholar]
- 15.Chen EY, Michel G, Zhou B, Dai F, Akhtar S, Schonberger RB. An Analysis of Anesthesia Induction Dosing in Female Older Adults. Drugs Aging. 2020;37(6):435–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Phillips AT, Deiner S, Mo Lin H, Andreopoulos E, Silverstein J, Levin MA. Propofol Use in the Elderly Population: Prevalence of Overdose and Association With 30-Day Mortality. Clin Ther. 2015;37(12):2676–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ni K, Cooter M, Gupta DK, Thomas J, Hopkins TJ, Miller TE, et al. Paradox of age: older patients receive higher age-adjusted minimum alveolar concentration fractions of volatile anaesthetics yet display higher bispectral index values. Br J Anaesth. 2019;123(3):288–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Van Cleve WC, Nair BG, Rooke GA. Associations Between Age and Dosing of Volatile Anesthetics in 2 Academic Hospitals. Anesthesia & Analgesia. 2015;121(3):645–51. [DOI] [PubMed] [Google Scholar]
- 19.Colquhoun DA, Shanks AM, Kapeles SR, Shah N, Saager L, Vaughn MT, et al. Considerations for Integration of Perioperative Electronic Health Records Across Institutions for Research and Quality Improvement: The Approach Taken by the Multicenter Perioperative Outcomes Group. Anesth Analg. 2020;130(5):1133–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kheterpal S Clinical Research Using an Information System: The Multicenter Perioperative Outcomes Group. Anesthesiology Clinics. 2011;29(3):377–88. [DOI] [PubMed] [Google Scholar]
- 21.Mathis MR, Dubovoy TZ, Caldwell MD, Engoren MC. Making Sense of Big Data to Improve Perioperative Care: Learning Health Systems and the Multicenter Perioperative Outcomes Group. Journal of Cardiothoracic & Vascular Anesthesia. 2020;34(3):582–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Larach MG, Klumpner TT, Brandom BW, Vaughn MT, Belani KG, Herlich A, et al. Succinylcholine Use and Dantrolene Availability for Malignant Hyperthermia Treatment: Database Analyses and Systematic Review. Anesthesiology. 2019;130(1):41–54. [DOI] [PubMed] [Google Scholar]
- 23.Lee LO, Bateman BT, Kheterpal S, Klumpner TT, Housey M, Aziz MF, et al. Risk of Epidural Hematoma after Neuraxial Techniques in Thrombocytopenic Parturients: A Report from the Multicenter Perioperative Outcomes Group. Anesthesiology. 2017;126(6):1053–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Benchimol EI, Smeeth L, Guttmann A, Harron K, Moher D, Petersen I, et al. The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement. PLoS Med. 2015;12(10):e1001885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Vandenbroucke JP, Elm Ev, Altman DG, Gøtzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. Annals of Internal Medicine. 2007;147(8):W-163–W-94. [DOI] [PubMed] [Google Scholar]
- 26.Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia in surgical mortality. JAMA. 1961;178(3):261–6. [DOI] [PubMed] [Google Scholar]
- 27.Schonberger RB, Dutton RP, Dai F. Is There Evidence for Systematic Upcoding of ASA Physical Status Coincident with Payer Incentives? A Regression Discontinuity Analysis of the National Anesthesia Clinical Outcomes Registry. Anesth Analg. 2016;122(1):243–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.WHO Expert Committee; Physical Status: The Use and Interpretation of Antrhopometry; WHO Technical Report Series #854; available at url: https://apps.who.int/iris/bitstream/handle/10665/37003/WHO_TRS_854.pdf;jsessionid=F930A944B6FD501F50902E08856DCCBB?sequence=1 as accessed October 5, 2020. 1995. [PubMed]
- 29.Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8–27. [DOI] [PubMed] [Google Scholar]
- 30.Multicenter Perioperative Outcomes Group Phenotype Browser; (url: https://phenotypes.mpog.org/ as accessed October 28, 2020).
- 31.van Walraven C, Austin PC, Jennings A, Quan H, Forster AJ. A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care. 2009;47(6):626–33. [DOI] [PubMed] [Google Scholar]
- 32.Sharma N, Schwendimann R, Endrich O, Ausserhofer D, Simon M. Comparing Charlson and Elixhauser comorbidity indices with different weightings to predict in-hospital mortality: an analysis of national inpatient data. BMC Health Services Research. 2021;21(1):13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Levey AS, Stevens LA. Estimating GFR Using the CKD Epidemiology Collaboration (CKD-EPI) Creatinine Equation: More Accurate GFR Estimates, Lower CKD Prevalence Estimates, and Better Risk Predictions. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2010;55(4):622–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16(1):31–41. [DOI] [PubMed] [Google Scholar]
- 35.Michels WM, Grootendorst DC, Verduijn M, Elliott EG, Dekker FW, Krediet RT. Performance of the Cockcroft-Gault, MDRD, and New CKD-EPI Formulas in Relation to GFR, Age, and Body Size. Clinical Journal of the American Society of Nephrology : CJASN. 2010;5(6):1003–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Annals of internal medicine. 2009;150(9):604–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Glance LG, Hasley S, Glantz JC, Stevens TP, Faden E, Kreso MA, et al. Measuring Childbirth Outcomes Using Administrative and Birth Certificate Data. Anesthesiology. 2019;131(2):238–53. [DOI] [PubMed] [Google Scholar]
- 38.Krumholz HM, Wang Y, Mattera JA, Wang Y, Han LF, Ingber MJ, et al. An Administrative Claims Model Suitable for Profiling Hospital Performance Based on 30-Day Mortality Rates Among Patients With Heart Failure. Circulation. 2006;113(13):1693–701. [DOI] [PubMed] [Google Scholar]
- 39.Natarajan A, Strandvik GF, Pattanayak R, Chakithandy S, Passalacqua AM, Lewis CM, et al. Effect of ethnicity on the hypnotic and cardiovascular characteristics of propofol induction. Anaesthesia. 2011;66(1):15–9. [DOI] [PubMed] [Google Scholar]
- 40.Kheterpal S, Vaughn MT, Dubovoy TZ, Shah NJ, Bash LD, Colquhoun DA, et al. Sugammadex versus Neostigmine for Reversal of Neuromuscular Blockade and Postoperative Pulmonary Complications (STRONGER): A Multicenter Matched Cohort Analysis. Anesthesiology. 2020;132(6):1371–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Mathis MR, Dubovoy TZ, Caldwell MD, Engoren MC. Making Sense of Big Data to Improve Perioperative Care: Learning Health Systems and the Multicenter Perioperative Outcomes Group. J Cardiothorac Vasc Anesth. 2020;34(3):582–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure S1: Histograms of Propofol Induction Dose by Age-Bin and ASA Physical Status. Vertical dotted line demonstrates the upper limit of the listed typical dosing range.
Supplemental Table S2: Table of Absolute and Weight Adjusted Propofol Induction Dose by Age and ASA Physical Status.
Supplemental Table S3: Table: Multivariable Analysis Evaluating Perioperative Factors Associated with Variability in Absolute Propofol Dose Received at Induction.
Supplemental Figure S4: Histogram of Absolute Propofol Dose in Milligrams for the Entire Dataset.