The authors should be congratulated for their efforts in ascertaining the immediate postoperative morbidity and mortality following spine surgery. They reiterate that adverse events can be frequent following hospitalization, anesthetic administration, and surgical intervention. While mortality was not necessarily related to the surgical intervention itself, it is sobering to learn that ten patients died within a month after surgery in this cohort of 3475 patients. Three patients developed a stroke, thirty patients in the same cohort developed wound infections, thirty-seven patients developed thromboembolic events, fifty-eight patients developed urinary tract infections, and 106 patients underwent repeat surgical intervention, all within thirty days of the index procedure. The authors’ list of adverse outcomes will contribute to our discussions with patients on the immediate risks of surgical intervention, promote more realistic patient expectations, and improve the process of informed consent. These data also provide additional information to policy makers, physicians, and hospital administrators in determining population incidences of adverse outcomes. This will help in formulating standards for patient care, "never events," and acceptable rates for hospital readmission.
The authors assert that this information obtained from the hospital-based National Surgical Quality Improvement Program (NSQIP) database will be more applicable to the American population as a whole than similar previously published data from a single surgeon or from one center. The database represents approximately 200 of the roughly 6000 hospitals in the United States, with a large proportion of Veterans Affairs and academic facilities. The availability of data on the cases of only 3475 patients who had spine surgery at 200 hospitals over a four-year period results in an average of four to five cases per year per hospital, and suggests substantial loss of data. While the incidences of morbidity and mortality are roughly similar to those published elsewhere, it is unclear that the volume and institutional representation of this database accurately reflect the breadth of surgical experiences in U.S. hospitals.
This study raises important issues on the design of an ideal registry for monitoring surgical outcomes. The NSQIP database is geared primarily toward the determination of mortality and morbidity in the thirty-day period following surgery. Ascertainment of harm in the perioperative period is an important part of the overall process of quality improvement from surgical procedures. Equally important to the process of quality improvement is longer-term survival, functional outcome, quality of life, pain relief, and comparison of different interventions. Registry data driven by surgeons and professional medical associations are likely to be more beneficial in developing the longer-term data that are critical to the informed decision-making process.
Where do we go from here? Postoperative adverse events are frequently linked by policy makers and patients with the surgeon and surgical procedure. This study clarifies that immediate postoperative complications are, for the most part, related to the age and preexisting medical condition of the patient. A better understanding of anticipated outcomes and their association with preexisting comorbidities will help patients to develop more realistic expectations from the procedure they are contemplating. Monitoring of anesthetic techniques and perioperative care processes will provide further clarification in ascertaining the causes of immediate perioperative morbidity and mortality, and eventually allow the development of better care pathways.
The incidence of adverse events in the immediate postoperative phase highlights the need for refined pathways of care for patients in the perioperative period. Algorithms or predictive models with thresholds of age or medical conditions that exclude patients from surgery are difficult to create, given the multiple variables involved. The surgeon bears ultimate responsibility in balancing the results of preoperative medical and nutritional evaluation and patient expectations from surgery, and deciding whether the benefits of surgery outweigh the associated risks. It is also important to recognize that surgery does not occur in a vacuum. Surgeons and hospital administrators should combine their efforts to ensure that effective multidisciplinary care, involving the anesthesiologist, internist, surgeon, nurses, and therapist, is provided throughout the perioperative period for the patient undergoing spinal surgery.