Introduction: There are several complications associated with spinal cord injury. The authors propose to evaluate the complications developed during hospitalization of tetraplegic patients treated in our institution.

Materials and Methods: The clinical and imaging records of 20 tetraplegic patients operated between 1995 and 2007 were evaluated (14 men and 6 women; mean age 31.5 years; 16 submitted to surgery using anterior cervical approach, 4 using posterior approach; 8 did steroids protocol during 24h and 12 during 48h; 9 patients were operated less than 48h after trauma and 11 patients after).

Results: Mean hospitalization time was 47.4 days (men 48.9 d, women 23.4 d; anterior approach 50.25 d, posterior approach 39 d; corticosteroids during 24h 34.3 d, 55.3 d in those who did 48h; time until surgery < 48h 43.1 d, > 48h 54.5 d). 100% of patients developed respiratory tract infections.

56.3% of patients developed urinary tract infections (33% in patients doing corticosteroids during 24h, 70% in those who did 48h)

Mean duration of mechanic ventilation was 20.3 days (anterior approach 19.3 d, posterior approach 19.8 d; steroids during 24h 16.7 d, steroids during 48h 21 d; time until surgery < 48h 13.6 d, > 48h 23 d)

In 37.5% of patients a traqueostomy was performed (41.7% in patients submitted to anterior approach, 25% in posterior approach; 16.7% in patients doing steroids during 24h, 50% in those who did 48h; time until surgery < 48h 28.6%, > 48h 50%)

Discussion: This patients are associated with long hospitalization and mechanic ventilation periods. Respiratory tract infection was the most frequent complication. The surgical approach had no influence on mechanic ventilation periods. Those submitted to anterior approach had longer hospitalization periods and higher incidence of traqueostomy. Patients who did corticosteroids during 48h had higher incidence of urinary tract infections and traqueostomy, and longer mechanic ventilation periods. Those operated less than 48h after trauma had shorter hospitalization and mechanic ventilation periods and traqueostomy procedure.

Conclusion: Steroids longer than 24h, anterior cervical approach and time to surgery > 48h tend to be associated with higher complication rates

A 14 year-old-female, underwent a T3-L3 instrumented posterior spinal fusion for a double major curve. Surgery under controlled hypotensive anesthesia was uneventful, with normal somatosensory and motor potentials. After instrumentation, patient underwent a normal wake-up test. The preoperative haemoglobin and haematocrit was 15.1g/dl with 41%, respectively. Estimated blood loss was 400cc and postoperative haemoglobin and haematocrit were 9.7g/dl and 31% respectively. Clinical examination was normal immediately postoperatively, on the first postoperative day and the beginning of the second postoperative day. At the end of POD 2, the patient started to feel both lower extremities “heavy” and sensitive to touch. She developed generalized proximal lower extremity weakness and was unable to stand. She was also unable to void after catheter removal. At this stage, her hemoglobin had dropped from 10 g/dl on POD 1 to 7.3 g/dl. Her haemoglobin fell to 6.2 g/dl the next day with a haematocrit of 18%. No significant bleeding was noticed, and other than lightheadedness, no haemodynamic changes were noted. Transfusion was performed correcting the haemoglobin to 9.3 g/dl and haematocrit to 27%. Compressive etiology was ruled out by post-operative myelogram-CT. Patient was discharged on POD 13 and was neurologically intact at three month follow-up. Discussion: Delayed neurological deficits have been reported, and are associated most frequently with epidural haematomas. Postoperative hypotension as the etiological factor has been reported only in an adult patient. As cord compression was ruled-out the only event we can correlate with the beginning of the neurological deficit is the unexplained acute drop in haemoglobin levels on the second day, possibly impairing normal cord oxygenation. If this is not the case, we would have to accept false negative results for the three standard methods currently available for spinal cord monitoring during surgery. In this case, the normal postoperative neurological exams, performed during the first 48 hours after surgery, and the subjective symptoms the patient experienced associated with the beginning of motor deficit, leads us to conclude that the injury happened on the second day in relation to the postoperative anaemia. Although we believe children tolerate low levels of haemoglobin, transfusion policies might have to be reconsidered as the cord may be transiently at risk for ischemic events after deformity correction.