US Deaths Resulting From Inpatient Surgery Decline

Janis C. Kelly

January 18, 2012

January 18, 2012 — The number of surgical procedures performed in the United States increased between 1996 and 2006, whereas inpatient postsurgical deaths within 30 days of admission dropped significantly, according to a national, population-level analysis published in the February issue of Surgery.

Marcus E. Semel, MD, MPH, from the Harvard School of Public Health and the Center for Surgery and Public Health at Brigham and Women's Hospital, Boston, Massachusetts, and colleagues used the Nationwide Inpatient Sample, which covers approximately 20% of hospitals in the United States, to examine the inpatient 30-day death rate for all surgical procedures, for procedures with the most deaths, for high-risk cardiovascular and cancer procedures, and for patients who suffered a recorded complication.

The researchers found that hospital admissions with surgical procedures increased from 12,573,331 in 1996 (95% confidence interval [CI], 12,560,171 - 12,586,491) to 14,333,993 in 2006 (95% CI, 14,320,983 - 14,347,002). Deaths within 30 days of admission dropped from 224,111 (95% CI, 221,912 - 226,310) to 189,690 (95% CI, 187,802 - 191,578) during the same period. The unadjusted 30-day inpatient mortality rate was 1.78% in 1996. After adjustment to match the sex and age of the 2006 population, the 1996 rate was 1.68%. In contrast, the inpatient 30-day mortality was 1.32% in 2006. By 2006, mortality had decreased significantly for 15 of the 21 procedures that had the most deaths in 1996.

The declines in inpatient deaths within 30 days of admission included patients at increased risk for poorer outcomes, those in the lowest income quartile, those with emergent admissions, those with a Charlson comorbidity index of 4 or greater, and those with Medicare.

The overall volume for the 21 principal procedures with the highest number of deaths in 1996 accounted for 3,160,643 admissions and 101,146 deaths. In 2006, these procedures accounted for 3,339,722 admissions and 74,254 deaths. Although the overall volume for these procedures increased, the number of open procedures declined. The volume of endoscopic and percutaneous procedures did increase, although the increase was not statistically significant.

Mortality decreased after 9 of 14 high-risk cardiovascular and cancer procedures. Mortality declined for all cardiovascular procedures except open abdominal aortic aneurysm repair, which declined in volume between 1996 and 2006 (volume, from 39,775 to 14,106 [P < .001]; mortality, from 11.19% to 11.15% [P = .970]). For cancer procedures, mortality declined significantly for pancreatic resection, esophagectomy, nephrectomy, and cystectomy.

In 2006, more patients had sepsis or pneumonia than in 1996, and the absolute number of deaths resulting from complications increased. However, the failure-to-rescue rate declined during the study period for both groups (sepsis, from 18.69% to 14.03%; pneumonia, from 8.54% to 7.34%). Meanwhile, the number of deaths resulting from deep venous thrombosis or pulmonary embolism, upper gastrointestinal bleeding, and shock declined during the study period.

Overall, the number of patients with 1 to 5 complications increased, but the failure-to-rescue rate for patients with an identified complication decreased from 12.10% to 9.84% (P < .001)

"The decline in the number of deaths may have occurred through reduced mortality of individual procedures, reductions in the volume of high-risk procedures, and the rescue of patients who had a complication," the authors write.

They estimate that 51,000 fewer people died in 2006 than would have with the 1996 mortality rate. However, they also note that some portion of the decline in mortality "may represent the effect of premature discharge as opposed to an actual improvement in survival." Data were not linked across admissions, and a patient discharged postoperatively who was later readmitted with a complication and died would not be counted as a death in this study.

The authors have disclosed no relevant financial relationships.

Surgery. 2012;151:171-182. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.