OTA 1998 Posters

The Risk of Multiple Organ Failure in Secondary Amputation following Limb Salvage in Multiple Traumatized Patients

Andreas Seekamp, MD; Gerd Regel, MD; Harald Tscherne, MD, Hannover Medical School, Hannover, Germany

Objective: The purpose of this retrospective analysis was to evaluate whether or not systemic parameters that are used to characterize multiple organ dysfunction would also indicate the risk of organ dysfunction in patients undergoing limb salvage despite a high mangled extremity score.

Methods: The principal criterium for the study group was a lower limb amputation following a type IIIb or IIIc open tibial shaft fracture in multiple traumatized patients. This group was then divided into one group of primary amputation (group A) and one group of secondary amputation (group B). Besides these groups, a third group of total traumatic lower limb amputaton was recruited (group C). Data analysis included demographics, injury severity according to the ISS, evaluation of the limb injury by two different salvage scores (MESS and NISSSA), and organ function monitoring by the Denver MOD Score over a 14 day period posttrauma or up to 7 days after secondary amputation.

Results: Within the period from 1987 to 1997, a total of 15 patients were recruited for group A (primary amputation), 10 patients for group B (secondary amputation) and 9 patients for group C (traumatic amputation). Primary amputation was performed in 15 patients of the grade IIIb and IIIc open fractures (group A, mean ISS 28.2). The main characteristics of these fractures were the vascular injury combined with the absence of plantar sensation. In the case of the two grade IIIb open fractures, primary amputation was performed due to the overall injury severity including the patient's age. Secondary amputation had become necessary in 10 patients (group B, mean ISS 21.0). In all cases, except one grade IIIb open fracture in this group, vascular injury, requiring repair, was evident, but plantar sensation was either positive or could not be evaluated initially. Mean time of limb ischemia was 6.2 hours. The decision for amputation was based on the presence of muscle necrosis and local infection. The third group of patients, all grade IV open fractures, received a completion of the traumatic amputation (group C, mean ISS 26.4). Not only vascular and nerve injury was evident but also skin and muscle were severely ruptured. Comparing all three groups regarding their clinical course after initial surgical treatment, no significant differences were noted at first sight. Although injury severity was significantly higher in group A compared to group B, mean time of intensive care was the longest in group B. According to the Denver MOD, score parameters for pulmonary, cardial, renal, and hepatic organ function were evaluated. A grade of organ dysfunction that would indicate an onset of multiple organ failure was only noted transiently in patients who did not survive in any of the three groups. Among all organ functions recorded for this analysis the pulmonary function and hepatic function were the only ones to reveal significant changes at all. Out of the parameters identifying pulmonary function, it was mainly the Horovitz quotient that showed any significant changes. For further illustration we have therefore focused on the Horovitz quotient and the Denver MOD score, which mainly was influenced by the pulmonary function. The impairment of pulmonary function was also evident simply by the time of ventilation. In group A (primary amputation), only 7 out of 15 patients required mechanical ventilation for more than 8 days. Only in these seven patients, pulmonary function decreased constantly according to the MOD score and had to be considered as grade 3 dysfunction for a mean period of seven days. The mean MOD score ranged between 3.0 and 3.33 from day 8 on. Three of these patients died due to pulmonary failure on day 17, day 18, and day 31 respectively. The other 8 patients of this group recovered within one week. In contrast, 8 out of 10 patients of group B (secondary amputation) required mechanical ventilation for more than 8 days. Despite this higher demand for mechanical ventilation, the MOD score and Horovitz quotient revealed only little deterioration of pulmonary function in this group of patients compared to group A. The mean MOD score of group B did not even reach a score of 2 (maximum was 1.57 at day four after amputation) at no time, except in the one patient who died. Pulmonary function was not even deteriorated significantly at the time of amputation, but on the other hand also after amputation pulmonary function did not fully recover as it was noted in the majority of patients after primary amputation. According to the mean MOD score in all patients, a grade 1 pulmonary dysfunction was evident before and after amputation. One patient of this group died on day 9, also due to pulmonary failure. In group C, only 5 of the 9 subtotal amputees received mechanical ventilation for more than 8 days, whereas all others had already recovered at day 3 after trauma. Pulmonary function in ventilated patients was not positive for severe dysfunction at any time, according to the mean MOD score and the Horovitz quotient. Only between day 8 and day 12 the MOD score indicated a grade 1 organ dysfunction in five patients (maximum 1.22 at day 8). In this group, all patients survived. As the ISS certainly has influenced the decision for amputation or limb salvage the results regarding organ dysfunction may be biased. Although the mean MOD score of primary amputees was significantly increased compared to secondary amputees, this could not implicate that secondary amputees are in general doing better than primarily amputated patients. For a most unbiased analysis, matching pairs of patients of either group were compared. The first pair (low ISS) included one patient of either group with an ISS of 15 and 14 respectively. Amputation was recommended in both patients according to the NISSSA but not the MESS. The primary amputee presented a MOD score of 1 throughout the six days of mechanical ventilation and was extubated on day 6. In contrast, the secondary amputee presented an organ dysfunction of grade 1 over 10 days and needed mechanical ventilation for a total of 13 days, amputation was performed on day 7. The second pair (high ISS) included two patients (mean ISS=31) of group A and one patient (ISS=33) of group B. In the first primary amputee, organ dysfunction of grade 1 was noted over the whole 14-day period except day 11 and day 14 on which it was raised up to grade 2. The second patient of group B revealed an increasing organ dysfunction up to grade 3 on day 6 and a return to normal on day 10. In contrast, the patient of group B, secondary amputee, presented a grade 2 organ dysfunction from day 6 on followed by an increase up to grade 3 until the end of the oberservation period. Amputation was performed on day 9. Again in both pairs of matching patients, the period of mechanical ventilation was significantly longer in secondary amputees.

Conclusion: According to our results, secondary amputation may lead to transiently decreased pulmonary function but does not necessarily end in multiple organ dysfunction. The right time for secondary amputation in order to prevent subsequent pulmonary dysfunction cannot be predicted by parameters otherwise indicating organ dysfunction. As the risk of secondary amputation for developing pulmonary dysfunction apparently cannot be estimated, the decision of amputation or limb salvage should be made initially after trauma and should be the definite one.