Surgical treatment of feline intracranial meningiomas: a retrospective study of 26 cases

Meningioma is the most frequently reported primary intracranial tumor in cats, representing 58% of brain tumors in this species [1, 2]. This slow-growing tumor arises from arachnoid meningeal cells [3]. Meningioma affects adult cats (3–21 years) [1, 4, 5, 6], and the median age at diagnosis is 11–12 years [2, 4, 7]. In cats, meningiomas are mostly solitary, but multiple meningiomas are not infrequent and may represent up to 17% of cases [1, 8, 9]. As feline meningioma is frequently located peripherally and is mostly non-invasive and benign (World Health Organization grade I), unlike canine meningiomas [1, 3, 5], surgical resection, when feasible, is often the treatment of choice. Other treatment options include radiation therapy or palliative therapy with or without chemotherapy [10]. Median survival time (MST) is 14 months [11] after radiation therapy and only 18 days after conservative treatment [1]. Surgery is associated with much higher reported survival rates (between 685 days and 37 months [1, 4, 7]), although tentorial location is associated with higher rates of recurrence and a less favourable prognosis [8].

Surgical treatment of intracranial tumors in small animals is also associated with relatively high perioperative and postoperative complication rates. Complications following craniotomy or craniectomy in dogs and cats usually occur in the early postoperative period: within 24 h in 35% of cases and between 1 and 10 days post-surgery in 52% of cases [12]. Between 32% and 47% of dogs experience major complications, such as death, seizures, worsening of neurological status and aspiration pneumonia, after surgical removal of intracranial tumors [12, 13, 14]. However, there are few reports investigating postoperative complications following meningioma excision in cats, nor reconstruction using cranioplasty. One study reported immediate postoperative complications and long-term recurrence [7], and another study investigated perioperative mortality [13]. A third study reported postoperative complications after brain surgery in dogs and cats [12].

Therefore, the aim of this study was to describe clinical findings, perioperative and postoperative complications, and patient outcomes in cats treated surgically for intracranial meningioma.

This retrospective study provides data about outcomes and complications after surgical resection of intracranial meningioma in 26 cats. In our retrospective study, MST of the whole study cohort was 881 days and was superior to the MST of one study [7] (655 days) but inferior to the MST of another study (1,113 days) [4]. However, the MST of cats that were still alive one month after surgery in our study was close to that found in another study (1,153 days vs. 1,143 days, respectively) [4]. No sex predilection was found in our study, contrary to what has been previously reported (1.1:1 vs. 1.8:1 [4] and 1.5:1 [1]). This could be explained by our small study population. Moreover, in our study, age was not associated with survival time, which is consistent with previously reported data [4, 7]. Therefore, surgical treatment should be considered in cats with intracranial meningioma regardless of age.

The peroperative mortality (0%) in our study was comparable to that of a previous report (2 of 121 cats) [4]. Death before discharge (11.5%) was lower than that reported in three previous studies, 18% [13], 19% [7], 17% [8], but higher than that of one study (6%) [4]. These discrepancies could be explained by variation in tumor location, which can complexify the surgical approach (among the three cats that died before discharge in our study, one had a combined approach) and differences in population size (type II error). Another contributing factor for our relatively lower rate of death before discharge may be the quality of postoperative care in the ICU between the study of Gordon et al. [7] and ours.

Meningioma recurrence is reported in 11%–39% of cats and occurs 3–60 months postoperatively [4, 7, 13]. Our results are consistent with these previous reports; in our study, 18% of cats had confirmed recurrence 19–37 months postoperatively.

The most common histologic subtypes identified in our study (transitional 39% and psammomatous 28%) were the same as those in other studies [4, 8]. However, fibrous and transitional meningiomas were more commonly reported in a previous study [5]. In our study, we did not examine the effect of meningioma subtype on outcome because all meningiomas were benign, and most of the cats still alive one month postoperatively died of causes unrelated to meningioma. Furthermore, no atypical or anaplastic meningioma was diagnosed in the cats in our study. High grade meningiomas are rare in cats. In previous studies, only five grade II and one grade III tumors were diagnosed histologically in 45 cases of feline meningioma (tumor obtained after surgery or necropsy) [5], and only one (anaplastic) meningioma was identified in 101 cases analysed after surgical excision [4].

Complications were not uncommon in our study (13/24, 54%). This complication rate is higher than those previously reported. In a previous study, 31% of cats had anaemia, 24% had no improvement or worsened neurological status, 14% had anorexia, and 10% had incisional drainage [7]. However, the overall complication rate was not reported in that study, and it is not clear whether some patients had multiple complications. In our study, complications occurred within 24 h after surgery in 92% (12/13) of cats with complications. Therefore, increased monitoring in the ICU (such as frequent clinical and neurological examination, Glasgow coma scale if relevant, continuous ECG, pulse oximetry, and invasive or non-invasive arterial pressure measurement) is highly recommended, at least for the first 24 h post-surgery to detect and manage possible complications and improve postoperative care. The only long-term complications reported in our study were consistent with tumor recurrence (altered consciousness (3/5), ataxia (2/5), weakness (2/5), circling (1/5)), which is similar to the results of a previous report of long-term complications in 10/33 cats [7].

Our study is the first to describe seizures as the most frequent complication in cats after meningioma resection (6/24; all male). Previous reports described anaemia as the most frequent complication following meningioma resection [7]. Seizures are a frequent complication after craniotomy or craniectomy in humans and dogs and may be related, among other causes, to manipulation or retraction of the brain parenchyma [15] and/or the presence of brain oedema [16, 17]. Meningioma size may also play a role in complication rate (and seizure occurrence). Increased manipulation of brain parenchyma may be needed to remove large meningiomas, despite the relative ease of dissection of feline meningiomas, compared with canine meningiomas. However, the size of the excised meningiomas was not always recorded in our study, and therefore, we could not analyse this factor.

A recent study reported early postoperative seizures (EPS) in 12.8% of neurosurgical surgeries and in 18.2% of dogs with rostrotentorial brain tumors after intracranial surgery [17]. The timing of EPS in that study is similar to the timing of postoperative seizures in the cats in our study (within 24 h after surgery). In human medicine, the incidence of EPS after supratentorial craniotomy for non‐traumatic pathology is estimated to be 1%–18% [18]. In a human study investigating EPS after brain tumor surgery, meningioma patients had the highest rate of EPS (10% vs. 5% for other types); in approximately 80% of these patients, a likely cause of the seizures, such as haemorrhage, increased oedema or meningitis, was identified on MRI. However, in human patients with meningioma, no correlation was found between tumor size (patients with postoperative seizures had bigger tumors but this difference was not significant), extra-axial haematoma, brain contusion, resection cavity haematoma, pneumocephalus and EPS [19]. In our study, no postoperative MRI or CT was performed in the cats that had postoperative seizures.

The relationship between EPS and preoperative seizures is unclear. While a study in dogs found no association between preoperative seizures and EPS [17], an association between the two has been found in human medicine [16]. In human patients undergoing resection of supratentorial meningioma, 22% had preoperative seizures. In this cohort, EPS occurred in 6% of patients during hospitalisation and seizures occurred in 14% of patients after discharge [16].

Tumor location may also be associated with post-surgical seizures. In veterinary medicine, rostrotentorial tumors [20, 21], especially within the frontal lobes, marked enhancement with gadolinium and subtentorial and/or subfalcine hernia were associated with seizures.

Moreover, sex has an influence on the incidence of seizures after meningioma surgery. In our study, the overall complication rate and the incidence of seizures after intracranial meningioma resection were significantly higher in male cats compared with female cats. In the humans, male sex is also associated with higher rates of seizures after meningioma resection [16] and higher rates of surgical site infections after craniotomy for tumor resection [22]. Despite this association, the role of sex hormones in the mechanism of postoperative seizures is still unknown. We were unable to elucidate the mechanism for this association in our study because we had a small study population, which may have resulted in a type II error, and four of the six male cats that had postoperative seizures were neutered. Therefore, this association warrants further investigation.

Postoperative anemia was not reported as a complication in our study. However, postoperative packed cell volumes were not recorded for all cats. Blood tests were not systematically performed if cats had no significant bleeding during surgery, were clinically alert and showed no sign of anaemia. Furthermore, no cat required blood transfusion before, during, or after surgery. However, this complication may have been underestimated, or information may have been lost because of the retrospective nature of this study. For example, one of the cats that died in immediate postoperative period had sinus bleeding during surgery and may have been anaemic.

In our study, only one septic complication was reported in a cat that had a cranioplasty. In humans, septic complications occur in around 8%–12% of cranioplasties [23, 24], and surgical site infection is the most common complication [24, 25]. However, in human medicine, cranioplasty is more often performed for decompressive surgery after stroke or traumatic brain injury (contaminated or bruised wound) than after an intracranial tumor resection. The use of an autologous bone flap is widely used for skull defect reconstruction in humans and the most frequent complications associated with these techniques are aseptic bone flap resorption and septic complications [26]. Synthetic materials, such as PMMA, hydroxyapatite, polyetheretherketone (PEEK), or titanium plates, are also used. These alloplastic materials are recommended because they are associated with less resorption and less infection [27]. PMMA is the most cost-effective synthetic material but is also associated with a greater infection rate than other synthetic implants. This is most likely because irregularities may promote bacterial adhesion, and biofilm formation and PMMA does not facilitate revascularisation [28].

In human patients undergoing tumor resection, cranioplasty is performed with an allograft or with synthetic biomaterial to avoid tumor recurrence when the tumor is in contact with or infiltrates the calvarium. Infiltration of the calvarium by feline meningioma is rare [29, 30], and hyperostosis is a frequent finding (up to 73%) [30]. However, in the authors’ experience, meningioma and the adjacent dura mater frequently adheres to the calvarium in cats. In a study of 48 human patients who underwent cranioplasty after tumor removal, eight (16.8%) had implant removal due to infection (surgical site infection in four, empyema in three and one with bacteraemia of unknown origin). The association between PMMA cranial vault prosthesis and surgical site infection needs to be investigated in small animals. In our study, cranioplasty was associated with a lower rate of complications. This may be because cranioplasty protects the brain parenchyma against external injury such as shock and against secondary compression by extracranial muscles, muscular bleeding or adherence formation. The physical barrier is not the only advantage of cranioplasty; it also helps to preserve cerebrospinal fluid and cerebral blood flow [31, 32]. However, this association may also be related to the fact that if an animal is not stable during surgery, the surgeon will decide to close the patient quickly and will not realise a cranioplasty.

In a previous study of dogs that underwent intracranial surgery (for brain biopsy, tumor removal or decompression after brain injury) only 2/50 had a cranioplasty with PMMA and 23/49 (47%) dogs had postoperative complications, including early postoperative neurological deterioration in 22/49 (45%) dogs [33]. Postoperative neurological degradation was not reported in our study, which may be because meningioma in cats is generally easier to resect than in dogs and may need less brain retraction and manipulation [3]. Cranioplasty may also have played a role in limiting this complication. However, postoperative neurological degradation may have occurred in cats that died in the immediate postoperative period and also may have been missed during recovery from anaesthesia. There also may have been a loss of information due to the retrospective nature of this study.

For the cats in our study, craniectomy site and surgical approach was chosen before surgery depending on tumor location and size. Measurements were taken on MRI or CT images to plan the surgical approach. Combined surgical approaches (a combination of at least two approaches that necessitates the occlusion of transverse venous sinuses or other large veins) were associated with more complications. In our study, one cat had transverse and sagittal sinus haemorrhage during surgery and died postoperatively after a bilateral rostrotentorial approach. Complex surgical approaches are more invasive, longer and harder to perform but may be required to remove large or invasive tumors. With the recent development of 3D preoperative planning, patient specific models and cutting guides, those complex approaches may become safer to perform [9]. Moreover, the use of neuronavigation may also reduce perioperative complications, but this was not available at our institution [34]. The use of operative microscope or endoscope assisted surgery may also help tumor dissection, limit local trauma to brain parenchyma and obtain clean tumor margins [35].

In our study a single surgeon performed all surgeries at a single centre, which may have limited the variability in patient outcomes and complications related to the surgeon. A surgeon’s experience can affect patient outcomes. For example, in human medicine, an association between surgeon technical skills and patient outcomes (complication rate, unplanned reoperation, death, or serious morbidity) has been reported in colectomy surgeries [36].

The main limitations of this study are the small patient population and the retrospective nature. As it was a retrospective study, we relied on the quality of existing medical records and therefore, complications may have been missed or underestimated. Nevertheless, it is unlikely that major perioperative and early postoperative complications were unrecognized because short-term follow-up was available in all cats. No cat had multiple meningioma in this study, which may represent a selection bias because those cats were less likely to undergo surgery. As multiple meningiomas represent approximately 17% of meningioma cases in cats [1], the lack of multiple meningiomas in our study may be due to the small patient population. A prospective study with a larger population is necessary to validate our results and increase the statistical power. However, the results of this study are similar to published data and provide further information regarding postoperative complications in cats after meningioma resection with or without cranioplasty.

This study confirms the benefits of surgical treatment of feline intracranial meningioma when tumor location permits resection. This treatment results in a long MST, and age at the time of surgery has no influence on outcomes. While complications were frequent, and the most common complication was EPS, complications may be limited by reconstruction with calvarial prosthesis. This information must be communicated to owners prior to surgery.