Predicting postoperative day 1 hematocrit levels after hysterectomy for malignant indication: Validating a previously...

Area of Science:

• Gynecologic Oncology
• Surgical Outcomes
• Laboratory Medicine

Background:

• The Swenson model predicts postoperative day 1 (POD1) hematocrit after benign hysterectomy.
• Its applicability to gynecologic oncology hysterectomies is not well-established.
• Accurate POD1 hematocrit prediction can optimize patient monitoring and resource allocation.

Purpose of the Study:

• To evaluate the accuracy of the Swenson model in predicting POD1 hematocrit in patients undergoing hysterectomy for gynecologic malignancies.
• To determine if the model can be reliably used as a screening tool in this population.

Main Methods:

• Retrospective data collection from 107 patients with malignant pathology undergoing hysterectomy (2014-2016).
• Comparison of predicted POD1 hematocrit (Swenson equation) with actual measured hematocrit.
• Receiver operating characteristic (ROC) curve analysis to identify optimal cut-off for predicting hematocrit ≤30%.

Main Results:

• The Swenson equation predicted hematocrit within ±5% for 83.2% of patients, showing lower accuracy than in benign cases.
• The model tended to underestimate lower hematocrit values, suggesting a conservative safety profile.
• ROC analysis identified a predicted hematocrit of 32.3% as the optimal cut-off for predicting actual hematocrit >30% with 100% specificity.

Conclusions:

• The Swenson equation demonstrates reduced accuracy for POD1 hematocrit prediction in gynecologic oncology hysterectomies.
• A predicted POD1 hematocrit of ≥32.3% suggests it is likely unnecessary to order postoperative hematocrit testing.
• The model can serve as a valuable screening tool to reduce routine laboratory utilization.