3.2 Study cohort
There were 631 patients with benign histology (either salivary or neurogenic in origin). The mean age (in years) at presentation was 42.9 ± 7.76 with a median follow up of 40.98 ± 19.1 months. Baseline characteristics and demographics along with presenting complaints, histological variation and complications of cumulative data have been further expressed (Table 1). The patients were categorized into two groups based on the tissue of origin such as neurogenic (n=310) and salivary (321), later compared on the basis of subsequent complications. The primary surgical approached used were (i) Trans-cervical (60%), (ii) Cervico-parotid (30%) including dissection and exposure of the facial nerve trunk and associated branches and (iii) Trans-mandibular (10%) comprising of mandible split for access and tumor resection (Figure 2). Trans-cervical approach was favored in 9 out of 13 studies as compared to 4 studies where cervico-parotid access was advocated. None of the studies has given clear preference to the use of trans-mandibular approach. The most common neurogenic tumor found was schwannoma (54%) followed by paraganglioma (43%) (Figure 2). The distribution of complications based on salivary or neurogenic histology has been represented for all the studies included in the review.
The 148 patients with neurogenic complication among 310 neurogenic tumors (Mean 11.38, SD 11.81) were compared against 41 patients with neurological complication among 321 salivary gland tumors and demonstrated significantly increased rate of nerve related complication (t = 2.42, p <0.05) (Figure 3). Among the neurologic complications, most encountered were vocal cords palsy (73%), Horner syndrome (9%), hypoglossal nerve injury (6%) and first bite syndrome (4%) (Table 2).
Factors considered for selecting surgical approaches in 13 studies on PPS tumors were categorized into size, location, histology, proximity to greater vessels or base of skull, histology and deep parotid lobe involvement (Table 2). Each factor was given a score and the combined scores were compared to the percentage of complications to assess if any of these factors considered before surgery would have an effect in the outcome in terms of complication rate. The correlation coefficient suggested a small degree of positive correlation that is not statistically significant (p=0.930) . The tissue of origin (neurogenic vs. salivary) seems to have a positive correlation with more complications related to neurogenic tumors (Pearson coefficient 0.662, p=0.019) (Figure 4). The binary logistic regression showed the significantly increased likelihood of developing post-operative neurological complications in tumors with neurogenic histology (p=0.0001) (Table 3).
Next we performed a meta-analysis assessing neurological complications among the groups having salivary or neurogenic tissue of origin. Due to heterogeneity < 50%, fixed term model was used. Data of 631 patients was available for analysis of whom 321 were of salivary tissue origin and 310 were neurogenic in histology. Neurological deficit was observed in 48% of patients with neurogenic histology (148/310) while only 13% patients with salivary tumor developed neurological deficit. The pooled RR was 2.41 (95% CI 1.80 – 3.23, p=0.001) (Figure 5).