"The take-home message is if you can do all the operation in one sitting it is better, [and] if you're able to avoid a second operation from the front, it is better."
Major perioperative complications occur in more than 8% of operations to correct adult spinal deformities, a research team led by Praveen Mummaneni, MD, of the University of California, San Francisco (UCSF), reported at the American Association of Neurological Surgeons 79th Annual Meeting.
The most common major complications were excessive blood loss, return to the operating room (OR) for deep infections, and pulmonary embolism (PE) or deep vein thrombosis (DVT). Dr. Mummaneni, associate professor of neurological surgery and codirector of spinal surgery at the UCSF Spine Center, said procedure-related risk factors that were controllable by the surgeon were the number of stages of surgery and the surgical approach.
In this eight-center, retrospective, case-control study, the researchers reviewed patient medical records until ten consecutive patients who met the inclusion criteria were identified at each site. A review of 953 medical records revealed 80 patients with major complications, eight of whom were excluded from the study because of biomechanical failures of implanted instrumentation. Therefore, the study included 72 cases with complications related to medical and operative factors. These cases were compared with a matched control group (N = 78) from each institution. Patients were operated on for the correction of coronal or sagittal spinal deformities.
The researchers reported 99 major complications (8.4%) in the 953-patient cohort, with an average of 1.4 complications per patient. Excessive blood loss greater than 4 L occurred in 11 patients, 11 patients were returned to the OR for deep wound infections, and 10 patients had a pulmonary embolus.
The proportion of single-stage operations was lower in the complications group compared with the control group (54% vs 63%, P = .011), and the complications group had more anterior-posterior surgical approaches (56.3% vs 32.1% for controls,P = .011), as opposed to posterior-only or anterior-only approaches. The length of hospital stay averaged 14.4 days for the complications group vs 7.9 days for the controls (P = .001).
Deep wound infections occurred mostly in revision operations, whereas DVT or PE was most common in the primary setting. Substantial blood loss (>4 L) was associated with more previous operations (P = .04) and longer surgical times (P = .018).
There were no significant differences between the complications group and the control group in terms of patient demographics, including sex, age, body mass index, number of comorbidities, American Society of Anesthesiologists grade, proportion of revision cases, or number of previous operations.
The researchers concluded that major perioperative complications were more related to factors attendant to the procedure than to factors related to the patients. This study did not look at minor complications.
"The interesting thing that we found was for patients who had these complications, the complications tend to be more if you did a front and a back operation or if you did operations on 2 different days," Dr. Mummaneni said. "And so for those kinds of patients the main complications we saw were DVT...or we saw infection, or we saw excessive blood loss, with prolonged ICU [intensive care unit] stay."
He added that patients who were having revision surgery were less likely to develop DVT, "which is the opposite of what we thought." Dr. Mummaneni recommended special caution during first operations and using a blood thinner postoperatively.
"The take-home message is if you can do all the operation in one sitting it is better, [and] if you're able to avoid a second operation from the front, it is better," Dr. Mummaneni advised.
Edward Benzel, MD, chairman of the Department of Neurological Surgery at the Cleveland Clinic in Cleveland, Ohio, called the collection and analysis of all the data a "herculean effort," and he congratulated the authors for "adding substance to the literature." He reminded the audience that the surgeons performing the operations in the study were very experienced. "These are very big operations," and the data should be interpreted accordingly, he said. He was not a participant in the study.
"My conclusion from this study is that bigger, longer cases are associated with a higher incidence of complications." He questioned whether the factors leading to more cases and longer operations are under the control of the surgeon or "is this a patient-specific pathoanatomic parameter?"
Dr. Mummaneni is a consultant to and has received other financial support from DePuy Spine, and he has received research support from Medtronic. Dr. Benzel reported financial or material support from Ortho MEMS, has served as a consultant to and received financial or material support from AxioMed, and has been a consultant to Spine Universe.
American Association of Neurological Surgeons (AANS) 79th Annual Meeting: Plenary Session I.
Daniel M. Keller, PhD • Medscape Medical News © 2011 WebMD, LLC