In-hospital outcomes of robotic versus laparoscopic inguinal hernia repair in obese patients: a national inpatient sample analysis 2005–2020

The National Inpatient Sample (NIS) is the largest publicly available all-payer inpatient care database in the United States (US). It is managed by the Healthcare Cost and Utilization Project (HCUP) and sponsored by the Agency for Healthcare Research and Quality (AHRQ). The NIS is designed to enable analyses of health policy issues at national, regional, and state levels. The database captures data from approximately 7 million hospital stays each year. After weighted, it estimates data of more than 35 million hospitalizations nationally, reflecting the US inpatient population at community hospitals. An overview of the NIS is available at: https://hcup-us.ahrq.gov/nisoverview.jsp)

The NIS database from 2005 to 2020 was examined for patients ≥ 18 years old who were obese (body mass index [BMI] ≥ 30 kg/m²), and underwent inguinal hernia repair by RIHR or LIHR. Patients with missing data on age, sex, sample weights, or outcomes of interest were excluded from the study. Diagnoses and procedures were identified using the International Classification of Diseases, 9th and 10th Revision (ICD-9 and ICD-10) codes, as summarized in Supplementary Table 1. Included patients were classified according to the types of surgery they received, RIHR or LIHR.

Patient characteristics such as age, sex, race/ethnicity, type of inguinal hernia (bilateral; unilateral), smoking status, and insurance status (primary payer) were recorded. Hospital-related characteristics, including weekend admission, admission type (emergent or elective), year of admission, bed number, location and teaching status, and region, were also extracted from the database. Major comorbidities (hypertension, diabetes, chronic pulmonary disease, chronic kidney disease, and rheumatic diseases) were considered as covariates, and the Charlson Comorbidity index (CCI) was calculated to assess the overall severity of comorbidities. Study outcomes included in-hospital mortality, length of stay (LOS), total hospital costs, and complications. The complications assessed were bleeding, urinary retention, infection, pneumonia, respiratory failure or mechanical ventilation, acute myocardial infarction (AMI), cerebrovascular accident (CVA), and venous thromboembolism (VTE). These conditions were also identified through relevant ICD codes, as documented in Supplementary Table 1.

This study was a secondary analysis of de-identified health claims data sourced from the HCUP. The utilization of anonymized administrative data for healthcare management and research purposes does not require specific written informed consent from the patient. Consequently, the ethical considerations associated with patient privacy and data confidentiality are inherently addressed through the use of these pre-existing, anonymized datasets.

The NIS database includes a 20% sample of US annual inpatient admissions, weighted samples (before 2011 using TRENDWT & after 2012 using DISCWT), stratum (NIS_STRATUM), and cluster (HOSPID) were used to produce national estimates for all analyses. The SURVEY procedure in SAS performs analysis for sample survey data. Patient descriptive statistics were presented as number (n) and weighted percentage (%), or mean and standard error (SE). Categorical data was analyzed by PROC SURVEYFREQ statement and continuous data was analyzed by PROC SURVEYREG statement. Logistic regression models were performed using the PROC SURVEYLOGISTIC statement to determine the associations between study variables and any complications, and data were reported as odds ratio (OR) and 95% confidence interval (CI). Linear regressions were performed using the PROC SURVEYREG statement to determine factors associated with LOS and total hospital cost. Multivariable regression was adjusted for significant variables (p < 0.05) in the univariate analysis. All p-values were 2-sided, and values of p < 0.05 were considered statistically significant. All statistical analyses were performed using the statistical software package SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA).

The patient selection process is illustrated in Fig. 1. A total of 650 patients ≥ 18 years old who were obese and underwent RIHR or LIHR were identified in the 2005 to 2020 NIS database. Three patients with missing information on total hospital costs and sample weight were excluded, leaving 647 patients in the study (representing 3,195 hospitalized patients in the whole US), of which 215 patients received RIHR.

The demographics, major comorbidities, and hospital-related information of the patients are summarized in Table 1. The mean age of the study population was 60.2 years and 78.0% were male. The RIHR group had a significantly higher proportion of males, private insurance coverage including HMOs, bilateral inguinal hernia, higher CCI scores, non-weekend admissions, non-emergent admissions, and admissions to hospitals with large bed sizes, urban teaching hospitals, and hospitals located in the Midwest compared to those who underwent LIHR.

Patients who underwent robotic surgery had significantly lower rates of any complications (16.3% vs. 22.3%, p = 0.019), shorter LOS (2.4 days vs. 3.5 days, p < 0.001), and higher total hospital costs ($78,000 vs. $58,100, p < 0.001) compared to those who underwent laparoscopic surgery. Additionally, patients who had robotic surgery experienced significantly lower rates of bleeding, urinary retention, pneumonia, and respiratory failure/mechanical ventilation (Table 1).

Multivariable analysis showed that patients who underwent robotic surgery had a significantly lower risk of any (at least one) complication (adjusted odds ratio [aOR] = 0.53, 95% confidence interval [CI]: 0.36 to 0.77, p < 0.001) compared to those who underwent laparoscopic surgery (Table 2). Additionally, robotic surgery was associated with a significantly shorter LOS (adjusted Beta [aBeta] = -0.94, 95% CI: -1.02 to -0.86, p < 0.001) and significantly higher total hospital costs (aBeta = 5.62, 95% CI: 4.80 to 6.44, p < 0.001) (Table 3).

Stratified analysis by age, CCI, hernia type, smoking status, hypertension, diabetes, and year of admission showed that robotic surgery was consistently associated with a significantly lower risk of complications in most subgroups. For patients < 60 years, the aOR was 0.56 (95% CI: 0.33–0.96), and for those ≥ 60 years, it was 0.53 (95% CI: 0.35–0.80). For patients with unilateral hernias, the aOR was 0.29 (95% CI: 0.17–0.52), indicating a significantly lower complication risk in the robotic group. Similarly, among patients with CCI ≥ 1, the robotic approach was associated with a reduced complication risk (aOR = 0.56, 95% CI: 0.37–0.84).

Robotic surgery was associated with a significantly lower complication risk during 2017–2020 (aOR = 0.50, 95% CI: 0.32–0.79). In earlier periods, it also showed a trend toward reduced complication risk, although this did not reach statistical significance (2005–2012: aOR = 0.33, 95% CI: 0.10–1.12; 2013–2016: aOR = 0.47, 95% CI: 0.19–1.15). (Table 4)

The major strength of this study is its use of a large, nationally representative dataset, which enhances the generalizability of the findings to the broader US population. The robust sample size allows for a comprehensive analysis of in-hospital outcomes, even for less common complications.

However, the study has several limitations. Most importantly, the retrospective design of this study carries a risk of potential selection bias, evident in the differing distributions of factors such as emergent admission status, hospital scale, and hospital teaching status, which may have influenced the outcomes. Although these factors were adjusted for in the multivariable analysis when found to be significantly associated with outcomes in the univariate analysis, the possibility of residual confounding remains. To address these limitations and provide more robust evidence, future randomized controlled trials (RCTs) are needed to validate these findings and overcome the inherent constraints of retrospective analyses when comparing the two procedures. Secondly, the reliance on ICD codes for identifying diagnoses and complications may lead to misclassification or underreporting of some conditions due to possible coding inaccuracies. Additionally, the higher cost associated with robotic surgery may not be feasible for all healthcare settings, limiting the applicability of the findings in resource-constrained environments. Moreover, the lack of data on important clinical variables such as intraoperative parameters and operation time limits the ability to fully assess and compare the procedural aspects of RIHR and LIHR. Lastly, the lack of long-term follow-up data precludes assessment of outcomes beyond the initial hospitalization period.