Survival rate and perioperative data of patients who have undergone hemipelvectomy: a retrospective case series

In this study, mean survival time was 32.8?±?4.6 months. In a previous study, Penna et al. [9] demonstrated a mean survival time of 43?±?17 months. This difference could be explained by the fact that a large number of patients had an advanced disease stage at the time of surgery in this study.

We found that the chance of survival was higher with internal hemipelvectomy than with external hemipelvectomy. This is consistent with the findings of a previous study [9]. Therefore, if possible, limb preservation surgery is the first choice when hemipelvectomy is considered a possible approach. The fact that external hemipelvectomy is currently performed in specific situations of more advanced disease such as failed neoadjuvant therapy, severe deep infection, sciatic nerve and femoral vessel infiltration, local tumor recurrence, improvement of the resection margin, and as a life-saving or palliative procedure could possibly explain the higher chances of survival in the internal hemipelvectomy group [2].

Although we did not notice any significant statistical difference in survival when patient age (50 or ?50) was considered (P?=?0.083), Mankin et al. [10] demonstrated a higher survival rate after hemipelvectomy in patients aged 50 than in those aged ?50. Sample size may have contributed to this difference.

Among the variables selected, only advanced disease stage was a significant predictor of reduced survival in both univariate and multivariate analysis (P?=?0.001), and this finding is consistent with the results of a previous study [10].

All patients with soft tissue tumors died at 60 months after the surgery in this study. However, Appfelstaedt et al. [8] reported a 10 % survival rate at 5 years for curative surgery and 14 % for palliation. On evaluating the data of patients with soft tissue tumors, we found that almost all the patients had an advanced disease stage (grade 3 or 4), and this might explain the reduced survival.

In the present series, 65.7 % of the procedures were external hemipelvectomy, although limb savage surgeries have been reported to be possible in most cases owing to medical advances [14]. Previous studies [9–12] have reported the predominance of internal hemipelvectomy in the treatment of pelvic cancer. This difference may be explained by the large tumor sizes in this study, and it reflects the high number of patients at an advanced disease stage assisted at Instituto Nacional de Câncer José Alencar Gomes da Silva.

Intraoperatively, a median of 2 units (range, 1–6 units) of RBCs were administered. Hemipelvectomy has been shown to be associated with massive bleeding and significant blood and fluid loss [15]. A previous study reported the requirement of massive transfusions [1], and another reported that a median of 7 units (range, 0–44 units) of RBCs were required intraoperatively [11]. Moreover, intraoperatively, a median crystalloid volume of 3500 mL (range, 1000–8500 mL) was infused, and the volume administered was lower in this series than in the study by Molnar et al. [11] (median, 8500 mL; range, 1000–42,000 mL). Additionally, the operation time was shorter than in the study by Molnar et al. [11] (200 vs. 300 min). These differences might be explained by the fact that none of the patients who underwent internal hemipelvectomy had bone reconstruction in this study, while 19 of 49 patients underwent some form of reconstruction in the study by Molnar et al. [11].

Balanced general anesthesia combined with epidural block was the most frequent choice of anesthesia for hemipelvectomy in this series, and this finding is similar to the finding of Molnar et al. [11]. Although some studies suggested that propofol anesthesia was superior to volatile agents in cancer patients [16, 17], other studies do not support this suggestion [18–20]. In a recent review, Heaney et al. [18] stated that there is no conclusive evidence to indicate that one anesthetic agent is better than another agent in cancer patients. Therefore, there is no reason to change the current practice.

We observed that efforts were made to deliver regional anesthetics in every case, and epidural or spinal local anesthetics were delivered in 32 of the 35 patients. Usually, after the surgical plan was discussed, epidural catheters were positioned in the lower thoracic or upper lumbar spine to avoid interference with the surgical field. A previous study reported that neuroaxial opioids were highly effective at reducing postoperative pain [2]. However, as 31.4 % of patients developed severe postoperative acute pain in our study, neuroaxial opioids were not able to provide adequate postoperative analgesia. Nevertheless, Weinbroum [21] reported the superiority of epidural over intravenous patient-controlled analgesia in orthopedic oncologic patients, and its use is recommended.

Chronic pain was an important postoperative complication. A total of 14 patients experienced this complication, and among the 23 patients who underwent external hemipelvectomy, 11 had persistent pain. Persistent postsurgical pain (pain that lasts for more than 1–2 months after surgery) has been reported to be present in more than 30 % of patients after procedures, such as amputations, and its recognition is increasing [22]. Studies have shown that up to 90 % of patients may experience phantom pain after hemipelvectomy-associated amputation, and although the mechanisms of pain after this procedure are not fully understood, the pain can be debilitating and may impair rehabilitation and quality of life [22, 23]. Neurotoxic chemotherapy, moderate-to-severe postoperative pain, anxiety, younger age (adults), radiation therapy in the operated area, and preoperative pain (moderate to severe) for more than 1 month were previously identified as risk factors for the development of chronic postsurgical pain [24, 25], and all or at least some of these factors may be present in hemipelvectomy patients.

In this study, 25,7 % of cases developed surgical wound complications that included infection, fistula, and dehiscence. Previous studies by Higinbotham et al. [26] and Apffelstaedt et al. [27, 28] reported wound complication incidences of 75 and 47 %, respectively. Beck et al. [22] reported a wound complication incidence of only 4 %. Our findings are somewhat similar to the findings of these previous studies.

The median postoperative length of hospital stay was 6 days (range, 3–27 days), and there was no difference between patients who underwent external hemipelvectomy and those who underwent internal hemipelvectomy. The length of time was shorter than in the study by Beck et al. [22] (14 days for internal and 20 for external hemipelvectomy) and by Molnar et al. [11] (14 days). Team efforts, institutional experience, and a low incidence of clinical complications may explain the earlier hospital discharge in our study.

This study did not focus on rehabilitation data. As a routine, all efforts were made to deliver early rehabilitation. Main goals were to sit in the first postoperative day and stand up in the fourth and, at discharge, patients were scheduled for ambulatory physiotherapy. Amputated patients received focus on wound edges care and preparation for prostheses. Internal hemipelvectomy patients received ambulatory ambulation training, corporal balance, and other mobility training such as gait aid.

A retrospective case series approach is useful to study rare diseases and infrequent procedures. However, several limitations are associated with this approach. This study was performed at a single center, and this might have resulted in bias. The large number of patients with advanced disease stage, especially in the soft tissue sarcoma group, may have contributed to low sample survival as well as worse outcome in external hemipelvectomy. Reports from different centers are important to contribute to increased knowledge about the outcomes of such aggressive and potentially critical procedures.