Objectives: To determine if a structured preoperative ICU consultation would correctly assign patients to preoperative invasive monitoring, postoperative ICU care, or recovery room care, and to compare morbidity, mortality, and resource utilization among all groups.
Design: Prospective, observational study.
Setting: A university hospital.
Patients: A total of 475 patients who were referred preoperatively by surgeons for ICU consultation and were evaluated by ICU physicians.
Interventions: Patients assessed to have clinical evidence of cardiovascular compromise were admitted preoperatively to the ICU for invasive hemodynamic monitoring and optimization. Patients without such evidence, but who were to undergo major operations or had anticipated major fluid replacement were independently selected for invasive monitoring by anesthesiologists. Patients who developed physiologic instability or became unstable due to hemorrhage also underwent invasive monitoring. Nonmonitored patients who remained stable were given postoperative ICU care or went to the recovery room based on an assessment by the surgeon and anesthesiologist at the end of the operation.
Measurements and main results: Of 8,916 elective surgical cases, ICU physicians were consulted in 475 (5.3%) patients preoperatively. Sixty-seven patients were admitted preoperatively to the ICU for invasive hemodynamic monitoring and optimization; 60 patients had surgery (0.7% of elective cases, 12.6% of ICU consultations). Patients selected for ICU preoperative monitoring were older than non-monitored patients and had higher numbers of cardiovascular and total risk factors than any other group. They had higher Acute Physiology and Chronic Health Evaluation (APACHE II) scores, higher Therapeutic Intervention Scoring System (TISS) points, a higher number of complications, and longer ICU stays than non-monitored postoperative ICU patients. In addition, they had a higher number of complications than nonmonitored recovery room patients. APACHE II scores, TISS points, number of complications, and ICU days in the preoperative ICU admission group were not increased when compared with all other monitored patients. Neither hospital days nor total hospital charges were increased when compared with the other elective ICU patients. Patients selected for ICU preoperative monitoring who underwent surgery had an 11.7% mortality rate and accounted for four of five cardiovascular-related deaths.
Conclusions: A small number of high-risk patients can be selected for preoperative monitoring on the basis of clinical assessment without increasing ICU stay or hospital bills. A structured preoperative consultation correctly identifies those patients who need monitoring and ICU care, but does not overutilize scarce and expensive ICU beds.