Introduction: Laparoscopic splenectomy is increasingly performed for both benign and malignant disorders. With advancements in laparoscopic techniques, the indications for laparoscopic splenectomy have broadened. To determine if clinical practice nationwide reflects the published superiority of the laparoscopic over the open approach, we compared the utilization of laparoscopic versus open splenectomy by analyzing the National Surgical Quality Improvement Program (NSQIP) dataset.
Methods: The NSQIP participant user file from 2005 through 2008 was used for data analysis. Pre-operative risk factors, intra-operative events, and post-operative complications were compared between the two groups. A multivariate model was constructed to evaluate the association between laparoscopic surgery, open surgery, and pre-operative risk factors with morbidity, mortality, and length of stay (LOS). ICD-9 codes were used to organize patients into five subgroups: (I) benign splenic disease, (II) malignancy associated with the spleen, (III) splenic injury, (IV) splenomegaly, and (V) miscellaneous diagnosis.
Results: CPT codes identified 2167 patients, of whom 48% (1045 patients) underwent laparoscopic and 52% (1122 patients) underwent open splenectomy. There was no significant change in the ratio of laparoscopic to open splenectomies as time progressed. Laparoscopic surgery was significantly (p<0.05) associated with younger age (51.6yo vs 56.9yo), higher BMI (28.3 vs 26.9), and female sex (55.6% vs 47.1%). In addition, laparoscopic patients were significantly less likely to have a history of COPD (3.5% vs 6.1%), renal failure (0.5% vs 1.4%), ascites (0.8% vs 4.5%), functional dependency on others (0.9% vs 1.4%), and to undergo emergent surgery (1.9% vs 26.5%). Subgroup analysis revealed that laparoscopic surgery was used more often than open surgery for group I (627 vs. 277 cases) and less often for group III (4 vs. 132 cases).
Overall morbidity rate was 20.6% (446 patients) with 28.9% in open and 11.7% in laparoscopic splenectomy. Overall mortality rate was 3.7% (81 patients) with 5.3% in open and 2.1% in laparoscopic splenectomy. With multivariate analysis, laparoscopic surgery was not significantly associated with a lower mortality rate but was significantly associated with lower morbidity and shorter LOS (p<0.001)
Excluding emergency cases, subgroup multivariate analysis revealed that laparoscopy was associated with a lower morbidity for groups I (OR 0.43, 95% CI 0.29 to 0.65) and group V (OR 0.28, 95% CI 0.12 to 0.67), but there was not a significant difference for either group II or IV. LOS continued to be significantly lower for laparoscopy in groups I-IV. There were too few laparoscopic cases in group III for meaningful analysis.
Conclusion: Despite literature proposing the superiority of laparoscopic over open splenectomy, less than half of the splenectomies identified in the NSQIP database were performed laparoscopically. While laparoscopic patients in this dataset were less likely to have certain risk factors, multivariate analysis demonstrates that there is still significantly less morbidity and shorter LOS associated with the laparoscopic approach, especially for those with benign splenic disease.AlthoughNSQIP data does not indicate the reason for laparoscopyversus open surgery, subset analysis by ICD-9 code and patient profiles suggest surgeon preferencerather than clinicalindications.
Session: Podium Presentation
Program Number: S004
« Return to SAGES 2010 abstract archive
