Abstract
Preoperative Anemia
We agree that anemia may contribute to postoperative complications, including venous thromboembolism. 1 As a result, we investigated the association of both blood loss anemia and chronic deficiency anemia and presented our findings in Supplementary Table 1A-1C. 2 Briefly, neither form of anemia demonstrated a significant association with lower extremity deep vein thrombosis (DVT) across any spinal level. Furthermore, the National Inpatient Sample (NIS) is limited to variables derived from diagnostic codes, in which the ICD-9 and ICD-10-CM classifications do not distinguish between mild, moderate, or severe anemia. While we did not observe a significant association, it remains possible that patients with mild anemia are grouped with those who have clinically significant anemia under the ICD-10 coding structure, potentially attenuating any underlying relationship. Thus, stratification by hemoglobin thresholds may identify a more nuanced relationship between hemoglobin levels and thrombosis risk across spinal levels. Currently, the NIS lacks laboratory values such as hemoglobin levels and does not provide temporal data to track hemoglobin trends over time. However, future investigations using databases with clinical laboratory values may clarify whether anemia severity confers differential risk of DVT by spinal level.
Obstructive Sleep Apnea
Obstructive sleep apnea (OSA) is an important risk factor for thromboembolic disease,3,4 and we appreciate the authors bringing attention to this point. Within administrative databases such as NIS, OSA is often underdiagnosed and inconsistently coded, increasing the potential for misclassification. 3 For this reason, we did not include it in our analyses. We agree that studies using datasets with verified clinical diagnoses or sleep study results would be better suited to examine the relationship between OSA and DVT in patients with spinal cord injury (SCI).
Intestinal Flora, Inflammation, and Small Intestinal Bacterial Overgrowth (SIBO)
The connection between gut dysbiosis, systemic inflammation, and thrombotic risk is an interesting and growing area of research.5-7 As mentioned previously, the NIS does not contain biomarkers, clinical measures of intestinal function, or indicators of SIBO, so this topic could not be addressed with the data available. However, the mechanisms described are biologically plausible, especially given the gastrointestinal dysregulation that accompanies SCI. We agree that this represents an important direction for future studies, particularly those designed with prospective or translational methods.
Closing Remarks
We are grateful to the authors for raising these points. While administrative databases allow us to study large, nationally representative samples, they also have clear limitations, especially for physiologic or mechanistic variables. We hope that future work combining large cohorts with more detailed clinical data will help refine our understanding of thromboembolic risk in SCI. We appreciate the opportunity to engage in this discussion and thank the authors again for their valuable feedback.
