In our study, 19.1% of the patients submitted to pheochromocytoma surgery experienced postsurgical complications, and 13% of them were classified as severe. Prolonged hypotension was the most common, occurring in 9.9% of the cases. Postsurgical complications were more common in elderly patients and patients with diabetes, a history of cerebrovascular disease, higher levels of plasma glucose and urinary free metanephrine and norepinephrine, and with pheochromocytomas larger than 5 cm.
The Clavien-Dindo classification was used for evaluating postoperative pheochromocytoma complications. The rate of postsurgical complications in our study was of 19.1%, but only 4 patients (13.0%) suffered from major complications (Clavien-Dindo ≥ 3). The rate of complications observed in our study is in accordance with other recent reported series from high-volume centres [7][8][9]. Several series also described prolonged hypotension as the most common postsurgical complications [15][7][16][17]. However, the rate of complications is widely variable among series, ranging from 10–40% [8][18][11][19]. There are several plausible explanations for this wide range, including referral biases, differences in the definition of complications and in the type of complications reported, and importantly, the decade of the performed research as it is known that a significant reduction in pheochromocytoma surgical treatment morbidity and mortality has taken place over recent decades [20]. In agreement with our study, series evaluating the severity of complications found similar figures of grade III and IV complications in the Clavien-Dindo score [10][9][11], although in some series severe complications occurred in up to 31% of the cases [8].
We found that postsurgical complications were more common in diabetic patients and in patients with cerebrovascular disease, as it has been previously described [8]. The higher risk for complications in elderly patients was related to the higher prevalence of these comorbidities. In line with our results, Srougi et al. [21] found that the mean Charlson score was 3.6 ± 1.3 in the older group and 0.89 ± 1.0 in the younger group, and elderly patients had a higher rate of postoperative complications (60% vs 18.9%, p = 0.01), but they did not adjusted the risk of complications for comorbidities as we did. Nevertheless, in a more recent study [22], the same authors included a larger number of patients, and they performed a multivariate analysis including age and comorbidities, observing that patients age (p = 0.004), comorbidities (p = 0.003) and pathological diagnosis (p = 0.003) were independent predictors of postoperative complications. In contrast with our results, a recent study comparing laparoscopic outcomes in elderly and young pheochromocytoma patients [23], did not find any difference in terms of haemodynamic changes during surgery and complications. However, this study included only 54 patients, and only two of them had experienced surgical complications, limiting the power of their results. No differences in hemodynamic instability, morbidity and mortality according to patient´s age [24][25][7][26][27][24] was reported in other studies either. So, based on results, although it is known that elderly patients usually have cardiovascular, pulmonary and hepatorenal changes that impact their response to anaesthesia, surgical stress and postoperative rehabilitation [28], the most important determinant of their higher risk of complications seems to be related to the association with higher rate of comorbidities in elderly, especially diabetes and cerebrovascular disease. Therefore, we suggest focusing on a strict control of comorbidities in elderly patients with pheochromocytoma undergoing surgery.
Other risk factor for postsurgical complications in our series was a higher level of urinary free metanephrine and norepinephrine, being urinary norepinephrine the most powerful biochemical predictor of complications. In accordance with our results, Namekawa et al. [16] showed that urinary epinephrine and norepinephrine were correlated with prolonged hypotension after tumour resection. Nevertheless, conflicting results have been described, some authors observed that urinary epinephrine (OR 1.18; 95% CI 1.04–1.35) and dopamine (OR 4.38; 95% CI 1.21–15.86), but not norepinephrine were independent risk factors for hypotension [17], and even some studies did not find any influence of the amount of catecholamines secretion and risk of postoperative hypotension [8][10].
We also observed that patients receiving selective alpha-blockade required intraoperative vasodilators more frequently than those treated with non-selective alpha-blockade. Similar results were reported in a recent systematic review and meta-analysis of 1344 patients [29], describing that intraoperative vasodilators were used more frequently in the group of selective alpha-blockade (OR: 2.46, 95% CI 1.44–4.20, P = 0.001). They also found a shorter length of hospital stay (WMD: -0.58 days, 95% CI: -1.12 to -0.04, P = 0.04) in patients treated with selective-alpha blockers, maybe the shorter stay could be related to the tendency to use phenoxybenzamine in hospitalized patients and use doxazosin on an outpatient basis. Supporting this finding, a recent randomized clinical trial comparing doxazosin and phenoxybenzamine [30] also described a need of higher number of vasodilating drug in the doxazosin group (P = 0.02). Nevertheless, they neither found differences in postoperative complications rate between both groups. The better control of intraoperative blood pressure, and the consequent lower need for vasodilators during surgery in the phenoxybenzamine group may be explained through its more effective inhibition of the α-adrenergic receptor, due to its non-competitive antagonism compared to the competitive binding provided by doxazosin [30].
Patients with tumours larger than 5 cm also had an increased risk of complications, but it was related to a greater functionality of the tumour. Tumour size is a well-known risk factor for postoperative [8][10][16][31][32][24] and intraoperative complications [11][24][27][33]. However, this relationship seemed to be due the real reason why size is associated with complications seemed to be related to a higher production of catecholamines in our study. In this line, Eisenhofer et al, [34] observed that tumour diameter showed strong positive relationships with 24-h urinary outputs of normetanephrine and metanephrine (r = 0.77; P < 0.001). Similar findings have been described by other authors [33], and also in other functioning adrenal tumours [35]. Besides, tumour size is a classical risk factor for recurrence [36] and malignancy [37]. Thus, tumour size should always be considered during the presurgical evaluation of pheochromocytoma and taken into account for a better planning of presurgical medical treatment and surgical approach.
Our study has certain limitations. First, it represents a retrospective review of data, which might incur recall bias. A second limitation is the likely variability of anaesthetic and surgical management between medical centres, The strengths of our study are the review of a large series of records from consecutive patients, with pheochromocytomas prepared for surgery in ten tertiary hospitals, and the precise definition of complications before data collection and analysis. Finally, a large sample of prospective randomized controlled studies is needed to further verify our conclusions.