Postoperative Outcomes Following Elective Surgery in India

DISCUSSION

This prospective cohort subset of international study has given us insights into patient characteristics, morbidity, and mortality following elective surgery in India. The observed complication and mortality rate following surgery in the Indian subset were 27.5% and 2.4%, respectively. This is considerably higher, compared to 19.8% and 0.5% in high-income countries (HICs) and 11% and 0.4% in low- and middle-income countries (LMICs), respectively.⁴ African Surgical Outcomes Study (ASOS) included data from 25 African countries undergoing all surgical procedures. They observed 13.4% complication and 1% mortality rate after elective surgery. However, surgical patients in ASOS were younger and had lower ASA grade.⁵ Therefore, LMICs and ASOS data should be interpreted with caution, as their patients were younger and healthier (ASA grade I and II). Hackett et al. evaluated the ability of ASA physical status to predict the postoperative complications and mortality in over 22 million patients in the American College of Surgeons ACS National Surgical Quality Improvement Program (ACS-NSQIP) in the 2012 database. They found that ASA physical status was a strong and independent predictive metric for predicting postoperative morbidity and mortality, across all types of surgeries. As the ASA physical status increased, so did the morbidity and mortality.⁶

The overall mortality in LMICs is similar to HICs despite low-risk patients, suggesting scarcity of resources in caring for the sick low-risk patients after developing complications. South African Surgical Outcomes Study (SASOS) replicated the EuSOS study and observed a crude mortality rate of 3.1% and included elective, urgent, and emergency procedures.⁷ Failure to rescue the patients after the development of complications in the Indian subset was 8.8%, which is considerably higher compared to 2.6%, 3.3%, and 4.8% in observed HICs, LMICs, and African nations, respectively.⁴,⁸ Failure to rescue is a well-accepted measure of postoperative quality of care.⁸ However, it should be interpreted with caution, especially when comparing data from different nations with varied health-care systems and case mix. Apart from a higher proportion of ASA grade III and IV patients, the Indian data set had a higher number (n = 132, 62.8% vs 20.3%) of cancer patients.⁴ Leeds et al. analyzed the 2005–2015 NSQIP data set of a gastrointestinal data set for the outcomes of patients undergoing surgeries for malignant versus benign lesions. They found that there was an increased likelihood of 30-day mortality (OR = 1.18, 95% CI = 1.10–1.28), complication rate (OR = 1.09, 95% CI = 1.07–1.11), and the total number of complications (OR = 1.08, 95% CI = 1.06–1.11) in patients undergoing surgery for cancer compared to benign lesions.⁹ More patients in the Indian subset underwent major surgical procedures (65.7% vs 36%), and this may have been an additional factor associated with increased complication rate. Indian patients with complications were older, with ASA grade III and IV physical status, and underwent major surgeries, which are known risk factors for postoperative complications. SASOS found age, ASA grade ≥II, major surgery, and infective etiology as an independent predictor of mortality.⁷ Another difference observed was a lower proportion of smokers in the Indian subset (6.8% vs 17.8%) compared to global data.⁴

In the Indian subset, 25% of patients were electively admitted to ICU postoperatively, compared to the 11.4%, 6.9%, and 7.8% observed in HICs, LMICs, and African nations, respectively.⁴,⁵ The higher rate observed in this subset may have been secondary to surgical unit protocols of caring for postoperatively in the ICU for 48 to 72 hours and/or until the availability of ward beds. Despite a higher rate of complications, 8.7% of patients were admitted to ICU for the treatment of complications compared to 3.2%, 2%, and 13.8% in HICs, LMICs, and African nations, respectively.⁴,⁵

Kahan et al. evaluated the ISOS data to study the impact of elective ICU admission on postoperative outcomes of patient, hospital, and national income levels. After adjusting for known confounders and exclusion of missing relevant data, they observed the rate of admission to ICU after complications increased to 15% and 17%, respectively, in both HICs and LMICs compared to that reported in the initial analysis.⁴,¹⁰ The observed mortality was higher in patients admitted to ICU compared to those managed on the wards (2% vs 0.3%). They attribute this difference to some residual confounders. At the hospital level, they found no association between mortality and elective ICU admission, ICU admission to treat complications. They report no survival benefit of elective postoperative admission to ICU.¹⁰ On the contrary, in the Indian subset, patients who were electively admitted to ICU were significantly older, a greater proportion had ASA grade III and IV physical status, underwent major surgery, and a higher proportion underwent cardiac and abdominal surgery. We observed that significantly more patients with elective admission developed complications (41.5% vs 22.7%). However, mortality in patients who developed complications was higher in patients who were not electively admitted to ICU [4/35 (11.4%) vs 1/22 (4.5%)]. Overall, there was no difference in mortality between the elective admission groups (2.6% vs 1.9%). The above results seem to justify the decision for elective postoperative admission of these patients to ICU. Conversely, intensive care beds are precious and limited resource in any hospital. Judicious use of this resource can help in lowering the failure to rescue rate. One aspect in the Indian context is the scarcity of trained nurses on the wards to care for these complex patients and therefore the need for establishment of protocols for elective postoperative ICU admission for 48–72 hours or longer based on surgical preference. Higher failure to rescue rate in the Indian subset may have been due to the patient and surgical factors and a possibility of nonavailability of critical care beds for critically ill patients, due to surgical protocols of caring for elective patients postoperatively irrespective of their risk stratification.

Abbott et al. reported the use of surgical checklist in 89.8% of patients in the ISOS data set. They found that exposure to surgical checklist was favorably associated with reduced mortality, but no difference in complications.¹¹ In the Indian subset, surgical checklist compliance was 80.7%. Similar to Abbott et al., there was no significant difference in complications despite increased compliance with surgical checklist in patients who developed complications.¹¹ In contrast to Abbott et al., we did not find any association between surgical checklist compliance and mortality.¹¹ These differences may have occurred due to patient characteristics such as older age, ASA grade II and IV, and major surgical procedures performed in our patients. The infection rate in our data set was more than twice the global rate (19.8% vs 8.9%), suggesting considerable scope for improvement and need for strengthening perioperative infection control measures. Chaudery et al. analyzed global ISOS data and reported 2% incidence of AKI in all patients and 4% following major surgery. We observed postoperative AKI was twice the global rate (3.9% vs 2%) for all patients and following major surgery (8.8% vs 4%).¹² This difference may have been due to older patients, with ASA grade III and IV, and different surgical case mix in the Indian subset compared to the global ISOS population. Similarly, we observed significantly higher mortality in patients who developed AKI. Cardiovascular complications were similar to the global rate (4.8% vs 4.5%).⁴ We identified preoperative hemoglobin as one of the independent predictors of complications, i.e., low hemoglobin was associated with a better outcome. We believe this to be a chance finding.

Based on the current study, one in three patients is likely to develop complications after elective surgery, and one in 11 patients is likely to die if they develop complications. These results are dismal compared to global data where one in six patients is likely to develop complications after elective surgery, and one in 35 patients is likely to die if they develop complications.⁴ Comparing data with other countries can help in setting benchmarks; however, various confounding factors such as case mix, ethnicity, differences in health-care systems, genetic predisposition, disease patterns, and access to health care should be taken into consideration. Surgery is a global health priority as per Lancet’s global surgery initiative (https://www.lancetglobalsurgery.org/india). Involvement of international organizations with national policymakers is likely to increase the surgical volumes in India. It is essential that future research should be directed in studying nationwide postoperative morbidity and mortality rates following elective surgery to establish a baseline and set benchmarks for improving outcomes and quality of care delivered.

There are certain limitations in the present study. Firstly, only three centers participated in this study, as a consequence Indian data was not analyzed for the primary outcome in the original study. Three centers cannot be considered to be representative of a nation like India with very diverse healthcare programs at the state level, availability of necessary facilities, skills and expertise, and access to health care in both government-funded and private hospitals. Nevertheless, it does give us some insights into patient characteristics and outcomes. Secondly, the data set collected limits our understanding of various other confounding factors within each hospital, type of institution (public or private), funded, or otherwise. Since much of healthcare expenditure is borne by the patients, there is a possibility that this can be an additional confounding factor for increased complications observed.