Secondary Logo

Journal Logo

Choice of Anesthesia for Cesarean Delivery: An Analysis of the National Anesthesia Clinical Outcomes Registry

Juang, Jeremy MD, PhD*†; Gabriel, Rodney A. MD; Dutton, Richard P. MD, MBA§; Palanisamy, Arvind MBBS, MD, FRCA*†; Urman, Richard D. MD, MBA*†

doi: 10.1213/ANE.0000000000001677
Obstetric Anesthesiology
Free

Neuraxial anesthesia use in cesarean deliveries (CDs) has been rising since the 1980s, whereas general anesthesia (GA) use has been declining. In this brief report we analyzed recent obstetric anesthesia practice patterns using National Anesthesiology Clinical Outcomes Registry data. Approximately 218,285 CD cases were identified between 2010 and 2015. GA was used in 5.8% of all CDs and 14.6% of emergent CDs. Higher rates of GA use were observed in CDs performed in university hospitals, after hours and on weekends, and on patients who were American Society of Anesthesiologists class III or higher and 18 years of age or younger.

Published ahead of print January 16, 2017.

From the *Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women’s Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts; Department of Anesthesiology, University of California, San Diego; and §US Anesthesia Partners, Dallas, Texas.

Published ahead of print January 16, 2017.

Funding: None.

Conflicts of Interest: See Disclosures at the end of the article.

Reprints will not be available from the authors.

Address correspondence to Richard D. Urman, MD, MBA, Department of Anesthesiology, Perioperative and Pain Medicine, Brigham and Women’s Hospital; and Harvard Medical School, 75 Francis St, CWN L1, Boston, MA 02115. Address e-mail to rurman@partners.org.

Historically, the type of anesthesia used during cesarean deliveries (CDs) has been an important determinant of maternal and fetal outcomes.1 In the United States, general anesthesia (GA) use has been steadily declining since the 1980s and is now largely supplanted by neuraxial anesthesia (NA).2–4 Single-institution and New York state data have shown continued declines in GA rates within at least the first decade of the 21st century.5,6 The recently published 30-year obstetric anesthesia workforce survey, inclusive of data up to 2011, showed a stably low rate of elective GA but an apparent rise in emergent GA use from 15% to 19% during the preceding decade.4

The National Anesthesiology Clinical Outcomes Registry (NACOR), a national repository for deidentified administrative, clinical, and quality-focused data formed in 2010 from anesthesia practices in the United States, allows easy electronic access to national anesthesia practice information at a lower cost and less effort than traditional methods for acquiring a similar breadth of data. Our goal in this brief report is to review anesthesia practice data for CD cases submitted to NACOR and to evaluate the feasibility of identifying facility and patient factors that may influence anesthesia choice.

Back to Top | Article Outline

METHODS

Data Source

This research was approved by our institutional review board and was exempted from the consent requirement. We analyzed records of CDs submitted to NACOR with current procedural terminology (CPT) codes 59510, 59514, 59515, 59618, 59620, or 59622 that occurred between January 1, 2010, and March 31, 2015. All CD cases were stratified by American Society of Anesthesiologists (ASA) physical status (PS) and the presence of E status, denoting emergency surgery. We note that ASA effectively eliminated the ASA I class for obstetric anesthesia in 2014, but the majority of the NACOR data predated that policy change.7 Additional case elements analyzed separately by all CDs included patient age, trial of labor after CD (TOLAC; CPT codes 59618, 59620, and 59622), year of surgery, time of surgery (7:00 am to 5:00 pm vs 5:01 pm to 6:59 am) as defined by anesthesia start time, day of week (weekend versus weekday), US region (Northeast, Midwest, South, or West), facility volume (1000 or fewer CDs versus >1000 CDs annually), setting (urban versus rural), and facility type. The facility type options included university hospitals, large community hospitals (consisting of >500 beds), medium community hospitals (100 to 500 beds), and other facility types, inclusive of small community hospitals (<100 beds) and specialty hospitals.

Back to Top | Article Outline

Statistical Analysis

R Project for Statistical Computing (R version 3.1.2) was used to perform all statistical analyses. We fitted multiple univariable logistic regression models to test whether each case characteristic, represented categorically, was associated with anesthesia type for all CDs. Results from the logistic regression were reported as odds ratios (ORs) with 95% confidence intervals (CIs). All characteristics that had a statistically significant association were used as covariates to fit one multivariable logistic regression model. A significance level of P < .0001 was used to avoid false positive findings in a multivariable model. Missing ASA PS values were imputed to a value of 2, and missing age was imputed to mean age of dataset (equal to 30 years). Missing values in nonordered categorical variables were coded as unknown. The area under the receiving operating characteristic curve (AUC) was calculated to estimate the performance of the regression model.

Back to Top | Article Outline

RESULTS

NACOR provided 287,127 CDs out of 26,568,734 total cases from January 2010 to March 2015.8,9 After exclusion of CD cases with unknown primary anesthesia type, the sample size was 218,285, with 205,671 (94.2%) cases performed under NA and 12,614 (5.8%) cases performed under GA. A total of 308 different hospital facilities contributed cases to this cohort, with a range from 1 to 14,385 reported cases per facility. A total of 15,282 emergent CD cases were identified, of which 13,046 (85.4%) cases were performed under NA and 2236 cases (14.6%) performed under GA. Temporal and total use of GA is depicted in the Figure.

Figure.

Figure.

Table 1.

Table 1.

Table 2.

Table 2.

Distribution of anesthesia rates (GA versus NA) by case characteristics for all CD cases is shown in Table 1. A summary of univariable and multivariable logistic regression analyses for all CD cases is shown in Table 2. The AUC for the logistic regression model was 0.645 (95% CI, 0.640–0.650).

Back to Top | Article Outline

DISCUSSION

The overall rate of GA usage nationwide comprised a small fraction (5.8%) of total anesthesia procedures performed during CDs. The GA rate at university hospitals was highest (8.5%) among facility types, an observation that was surprising because the GA usage rates spanning from 2000 to 2005 published previously based on a single university hospital ranged from 0.4% to 1%.6 The rate of GA usage during emergent CD reported in the NACOR dataset (14.5%) is consistent with more recently published work by the obstetric anesthesia work force survey.4 Interestingly, the 30-year update from the workforce survey reported a slight increase in emergent GA use from 15% to 19%; however, the statistical significance of this rate increase is not discussed in that article.

Although NACOR is expected to capture at least basic data from approximately 25% of all anesthesia cases in the United States in 2015,10 the number of CD cases included over the period of January 2010 to March 2015 (287,127) appears to represent a small proportion of the estimated number of CDs performed in the United States during the same period (6.8 million).11 The annual number of CD cases in NACOR more than doubled between 2010 and 2014, indicating evolving reporting practices. These observations suggest that although the NACOR data allow for the identification of subsets of data elements that are associated with higher rates of GA use, our statistical analysis can only be considered illustrative; results may change as the sample collected by NACOR expands. One observation supporting the validity of the NACOR dataset is that the derived emergent GA rate is similar to obstetric anesthesia work force survey results from the preceding decade.4

A few additional methodologic limitations warrant discussion. NACOR does not allow us to identify subsets of GA cases resulting from NA failure, which can vary widely between institutions (2% to 20%).12–14 NA placed for vaginal deliveries that were subsequently converted to CDs also cannot be distinguished from NA placed de novo for CDs. Consequently, the relationship between institutional characteristics and use of GA is likely confounded by the unmeasured relative contributions of elective CD and unplanned CD in each institution. For CD-type designations, NACOR also does not explicitly enforce a standard definition for the emergent classification. The American Congress of Obstetricians and Gynecologists and American Academy of Pediatrics both define emergent as the interval from decision to incision of <30 minutes; however, minor variations to the 30-minute rule have been reported across practices. We did observe from the NACOR dataset that emergent status increases the likelihood of GA use when TOLAC decreases the likelihood of GA use.

In summary, NACOR is the largest national database for anesthesia practice information and the best available data source to help identify current practice patterns, outcomes, and eventually longitudinal practice trends. Future studies will be needed to validate our observations against independent national data sources.

Back to Top | Article Outline

DISCLOSURES

Name: Jeremy Juang, MD, PhD.

Contribution: This author helped design and conduct the study, and helped with data collection and analysis, and manuscript preparation.

Conflicts of Interest: None.

Name: Rodney A. Gabriel, MD.

Contribution: This author helped design and conduct the study, and helped with data collection, analysis, and manuscript preparation.

Conflicts of Interest: None.

Name: Richard P. Dutton, MD, MBA.

Contribution: This author helped with data analysis and manuscript preparation.

Conflicts of Interest: Richard P. Dutton is a member of the Data Use Committee, Anesthesia Quality Institute.

Name: Arvind Palanisamy, MBBS, MD, FRCA.

Contribution: This author helped with data analysis and manuscript preparation.

Conflicts of Interest: None.

Name: Richard D. Urman, MD, MBA.

Contribution: This author helped design and conduct the study, and helped with data collection, analysis, and manuscript preparation.

Conflicts of Interest: Richard D. Urman is a member of the Data Use Committee, Anesthesia Quality Institute.

This manuscript was handled by: Jill M. Mhyre, MD.

Back to Top | Article Outline

REFERENCES

1. Andersen HF, Auster GH, Marx GF, Merkatz IR. Neonatal status in relation to incision intervals, obstetric factors, and anesthesia at cesarean delivery. Am J Perinatol. 1987;4:279–283.
2. Hawkins JL, Gibbs CP, Orleans M, Martin-Salvaj G, Beaty B. Obstetric anesthesia work force survey, 1981 versus 1992. Anesthesiology. 1997;87:135–143.
3. Bucklin BA, Hawkins JL, Anderson JR, Ullrich FA. Obstetric anesthesia workforce survey: twenty-year update. Anesthesiology. 2005;103:645–653.
4. Traynor AJ, Aragon M, Ghosh D, et al. Obstetric anesthesia workforce survey: a 30-year update. Anesth Analg. 2016;122:1939–1946.
5. Guglielminotti J, Wong CA, Landau R, Li G. Temporal trends in anesthesia-related adverse events in cesarean deliveries, New York State, 2003-2012. Anesthesiology. 2015;123:1013–1023.
6. Palanisamy A, Mitani AA, Tsen LC. General anesthesia for cesarean delivery at a tertiary care hospital from 2000 to 2005: a retrospective analysis and 10-year update. Int J Obstet Anesth. 2011;20:10–16.
7. Practice guidelines for obstetric anesthesia: an updated report by the American Society of Anesthesiologists Task Force on Obstetric Anesthesia and the Society for Obstetric Anesthesia and Perinatology. Anesthesiology. 2016;124:270–300.
8. National Anesthesiology Clinical Outcomes Registry. Anesthesia Quality Institute. Available at: https://www.aqihq.org/introduction-to-nacor.aspx. Accessed November 17, 2016.
9. McIntosh CA, Macario A. Managing quality in an anesthesia department. Curr Opin Anaesthesiol. 2009;22:223–231.
10. Liau A, Havidich JE, Onega T, Dutton RP. The National Anesthesia Clinical Outcomes Registry. Anesth Analg. 2015;121:1604–1610.
11. National Center for Health Statistics Centers for Disease Control and Prevention. 2016 Birth Data. Available at: http://www.cdc.gov/nchs/fastats/delivery.htm. Accessed May 4, 2016.
12. Lee S, Lew E, Lim Y, Sia AT. Failure of augmentation of labor epidural analgesia for intrapartum cesarean delivery: a retrospective review. Anesth Analg. 2009;108:252–254.
13. Orbach-Zinger S, Friedman L, Avramovich A, et al. Risk factors for failure to extend labor epidural analgesia to epidural anesthesia for Cesarean section. Acta Anaesthesiol Scand. 2006;50:793–797.
14. Campbell DC, Tran T. Conversion of epidural labour analgesia to epidural anesthesia for intrapartum Cesarean delivery. Can J Anaesth. 2009;56:19–26.
Copyright © 2017 International Anesthesia Research Society