In the present study, timing of intervention for acute, severe, symptomatic pituitary apoplexy was found to be predictive of visual function recovery at only the 7-day timepoint, as has been reported by previous studies [18, 44, 45]. However, no significant difference in post-operative visual outcomes was found when comparing interventions performed prior to versus after the 48 h or 72 h timepoints. Ultimately, this suggests that no further optimization of outcomes is obtained by taking PA patients to surgery “as soon as possible”. Nonetheless, drawing upon multiple studies that have been published over recent decades, operating within the first week does appear to afford superior vision outcomes when compared to “late” interventions (operating at > 7 days beyond the onset of apoplexy).

Furthermore, surgical decompression and conservative management appear to be effective treatments for pituitary apoplexy. Neither demonstrates superiority when assessing for ophthalmoplegia or cranial nerve deficits, visual field testing, or visual acuity [45]. Previously, reports have suggested that initiating conservative management is indicated when symptoms are mild at onset, but that severe visual deficits warrant emergent surgical decompression of the pituitary gland [11, 28]. Thus, both interventions can be effective when selected on a patient-to-patient basis. Ultimately, this is consistent with the results of the present meta-analysis. Nonetheless, since the 1970s, reports have described spontaneous recovery and remission of hyperpituitarism in PA, citing conservative management as instrumental to the successful outcomes obtained [21, 46, 47]. Eventually, this practice ascended to become the first-line treatment for severe PA at multiple pituitary centers worldwide [45]. Simultaneously, others have argued that surgical intervention should be automatically performed for patients demonstrating any acute onset neurological deficits [45].

Ultimately, the present analysis confirms that the standard-of-care management of PA has shifted over the decades from “surgical emergency” (take to OR immediately) to surgically urgent (< 48 h) to the present trend in care: surgery within 7 days or treatment with conservative management in patients with mild deficits. At present, the precise timing of surgery appears to be surgeon specific and training-dependent. In other words, preferences may be experience-dependent as well as based on the acuity and severity of symptoms, tumor size, histopathology, and secretory function of the tumor. All things considered, the present manuscript is intended to provide guidance for the subacute management of patients who develop this condition, as most cases will likely lie somewhere between clearly emergent (meriting immediate surgery) and non-emergent apoplexy.

Perspectives on management of pituitary apoplexy over the decades

Clearly, management continues to represent an ongoing source of debate, and thus some skull base surgeons have taken a midline stance, choosing to tailor their approach to the specific patient. For example, some reserve surgery for patients with acute vision loss or other neurological deficits and manage patients with mild symptoms conservatively [45]. Even when these patients are treated conservatively, elective pituitary surgery remains an option for them down the line. Though simple in concept, implementing this treatment paradigm in reality is nuanced and not always straightforward. Most PA patients typically present somewhere along a spectrum ranging between the two categories: 1) mild, benign and 2) acute, emergent. As a result, many patients experiencing apoplexy contain hemorrhagic pituitary lesions that can be managed conservatively. However, some might argue that these patients should be taken to surgery [45].

To facilitate clinical decision-making, it can be helpful to reference a pituitary apoplexy grading scale. One such example is the Pituitary Apoplexy Score (PAS), which has previously been validated by Bujawansa and colleagues [28]. The PAS assigns points for diminished level of consciousness (GCS 15 = 0, GCS 2–7 = 4), visual acuity (normal = 0, reduced-unilateral = 1, reduced-bilateral = 2), visual field deficits (similar to visual acuity scoring), and ocular paresis (scored from 0–2 in a manner similar to visual acuity and visual field) [45]. One issue with this scale is that it does not necessarily provide precise quantification of deficits in mild forms of PA; as such, Jho and colleagues designed a modified Pituitary Apoplexy Grading System that simplifies grading into a scale ranging from 1–5. A score of “1” corresponds to absence of symptoms, while a score of “2” designates endocrinopathy only. Any visual acuity loss or visual field deficit is automatically assigned a score of 5, as is any patient presenting with a GCS too low for testing. The benefit of the Pituitary Apoplexy Grading System is that it is more sensitive for detection of mild symptomatology and enables a more straightforward indication-based assessment of visual function and overall neurologic status.

United Kingdom guidelines for the management of pituitary apoplexy [48]

Currently, one of the most referenced guidelines for the treatment of PA are the U.K. guidelines developed by Wass and colleagues of the University of Oxford’s Churchill Hospital [48]. As stated within the guidelines, PA is a rare emergency most often associated with clinically nonfunctioning macroadenoma [11, 49]. Greater than three-quarters of the time, it is the first indication that a patient has an underlying, nonfunctioning pituitary macroadenoma. The condition is associated with a slight male predominance of approximately 1.6:1, and indeed the majority of patients in the current study were male [50, 51].

Because PA is often acute in onset and the first sign of an underlying pituitary lesion, diagnosis and intervention must proceed in a timely manner, but the reality is that the clinical picture can be conflated by uncertainty. For example, initial management is often dependent on the health care setting patients initially present to. In various settings, access to subspecialty opinions from endocrine, neurosurgical, and ophthalmologic providers may not be readily available [48]. When taken together with the fact that early versus late surgical decompression for PA is a topic of great debate, the waters surrounding optimal management of PA remain murky. It is thus not entirely clear which approach is the best for minimizing morbidity and mortality.

The underlying etiologies of the visual deficits associated with pituitary apoplexy – aside from optic nerve compression – are similar to deficits observed in cavernous sinus syndrome. Deficits are the result of sudden hemorrhage that results in rapid accumulation of blood in the confined space of the sella turcica and cavernous sinus, explaining why emergent alleviation of rapid bleeds is referred to as surgical decompression of the pituitary and its surrounding structures.

Type of surgical intervention

Compression of cranial nerves III and VI can result in ophthalmoplegia, an extraocular muscle palsy that has been reported in 70% of cases of pituitary apoplexy [52, 53]. Of the cranial nerve palsies observed in apoplexy, CN III palsy is in fact the most common and represents approximately half [49]. When mass effect is exerted upon the optic chiasm, bitemporal hemianopsia will be a presenting sign/symptom – reported in approximately 75% of pituitary apoplexy patients with visual dysfunction [41]. Prior literature suggests that patients with mild visual deficits are candidates for conservative treatment and that ophthalmoplegia will resolve without surgical intervention in these cases [54]. Nonetheless, exceptions to this rule can arise and this is only reasonable management for patients with an overall clinical presentation that can be considered mild [24].

When permissible, conservative treatment is preferred because surgical decompression is associated with risks of any surgical procedure involving the pituitary gland: endocrinopathy, hemorrhage, CSF leak, and death [11]. Performing surgery for pituitary apoplexy is indicated when significant cranial nerve and/or ophthalmologic defects are detected on exam or the patient presents in or deteriorates into a low state of consciousness/mental status (indicating significant mass effect or elevated ICP). Altogether, our analysis suggests that, among studies claiming to adhere to treatment protocols along these lines, outcomes are roughly equivocal following conservative management and surgery. Both can be effective when patient selection is performed carefully and dictated by the results of a thorough, meticulous neurological examination and subsequent documentation of any deviations from the baseline exam [38].

Timing of surgical intervention

Of course, management of pituitary apoplexy is not quite this simple: seeing this through is easier said than done, no single patient is alike, and like most pathologies of the CNS, management can be nuanced, and gray areas do exist. As previously mentioned, the seven-day window is generally considered the ideal timeframe within which to operate, and this recommendation is supported by the U.K. guidelines [48]. Despite this being a commonly respected recommendation, previous studies have found that visual field and acuity outcomes before and after the seven-day timepoint are comparable; however, this does not appear to be the case for patients with preoperative ophthalmoplegia. In fact, these patients experience ocular palsy recovery rates of roughly 30%-60%, far lower than ocular functional recovery rates observed when ophthalmoplegia is absent (70%-90%) [13, 24, 28]. This suggests that ophthalmoplegia is a sign indicating that more urgent surgical intervention is warranted.

Regardless of timing, it is essential that fluid resuscitation and possibly corticosteroids be administered in any severe presentation that is slated to undergo surgery [53]. PA patients should be stabilized through monitoring of electrolyte levels and managing pituitary hormone imbalances (such as administering hydrocortisone to prevent circulatory collapse) [8]. Furthermore, the PAS grading system has emerged as a useful tool that can help neurosurgeons decide whether to intervene operatively (after the patient is stabilized). The PAS, introduced by Reddy and colleagues, offers a method for deriving a calculated measure of the severity of a given PA presentation, and several retrospective studies have suggested that it is useful in predicting whether conservative versus surgical management should be pursued [28, 55, 56]. Further investigations are merited so that the validity of this tool can be verified for clinical application.

Recent updates regarding treatment of pituitary apoplexy

In 2021, Shepard and colleagues received attention after publishing their single-center retrospective study on clinical and radiologic outcomes of pituitary apoplexy by management strategy (conservative versus early surgical intervention) [7]. The authors reported that most PA cases can be successfully managed with conservative treatment, particularly when patients present with minimal visual defects such as incomplete bilateral temporal hemianopia or partial cranial neuropathy). However, as pointed out by Wang and colleagues, early surgery was not categorized on the basis of what would typically be defined as “early” (< 48 h) [57]. Instead, the authors defined early surgery as surgery within 1 week of PA diagnosis. The study’s findings, that surgery should be performed within 1 week for patients with severe visual field deficits and ocular palsies, are therefore consistent with recommendations offered in the literature. With respect to the surgery versus conservative treatment debate, patients who underwent surgery in Shepard’s study had larger PA volumes and tumor diameters, on average, than patients in the conservative group. Coupled with the finding that patients in the early surgical group had larger tumors on average, Shepard reported that patients with deteriorating vision and/or severe visual impairment were more likely to undergo early surgery [7]. The difficulty in interpreting these results is that it is difficult to capture what role initial presentation had in ultimate prognosis. In other words, were ultimate outcomes influenced more by severity of presentation or the course of treatment pursued (surgical versus conservative)? In response to this question, Shepard and Jane Jr. acknowledged that management of PA remains a highly controversial topic. Furthermore, they reiterated that their series was meant to illustrate their institutional practice for management of patients with PA: conservative treatment with high-dose steroids and close neurologic/ophthalmologic monitoring, and that their goal was to supply evidence indicating that successful outcomes can be obtained through conservative management of PA, indicating that it is not universally a condition that merits emergency surgery [7, 58]. Furthermore, Shepard and Jane Jr. emphasized that all patients with normal visual function were managed conservatively, and that there were no cases of visual function decline in these patients. Of patients in whom conservative management failed, only three had transient decline in visual function that improved following surgery. By contrast, all patients who underwent early surgery for visual acuity or field deficits (or ophthalmoplegia) in the setting of PA experienced subsequent visual decline. Though their outcomes were worse, Shepard and Jane Jr. argue that this is because early surgery should be reserved for the most severe cases of PA; the cases that by default will exhibit more advanced deficits post-intervention (as this is further proof of the urgent and/or emergent nature of their presentation). Otherwise, Shepard and Jane Jr. suggest that conservative management is effective for non-severe cases of apoplexy and that the challenge that has hindered the development of one standard guideline for treatment of PA is that apoplexy is a spectrum of diseases as opposed to a single clinical entity. Finally, they acknowledge that early surgery may be reserved for select cases in which patients present with severe visual acuity loss or visual field defects, but that the majority of cases can be managed conservatively. When surgical intervention is indicated, there is no evidence that performing surgery < 48 h following apoplexy provides any advantage over intervening < 1 week after onset. Accordingly, multiple reports within the current literature consider any surgery performed within 1 week as “early” surgical intervention [7].

Finally, another relevant question that has emerged is whether or not outcomes differ depending on whether or not the pituitary tumor is a functioning or non-functioning adenoma. At present, the literature indicates that there is little to no difference in outcomes between patients experiencing apoplexy who harbor functioning adenomas as compared to those who have underlying non-functioning adenomas [59]. Of course, the vast majority of patients with apoplexy secondary to pituitary macroadenoma experience apoplexy as the first sign of their underlying pituitary mass. If not for the apoplexy, the pituitary lesion would continue growing in an asymptomatic manner until becoming symptomatic due to mass effect. Because non-functioning pituitary adenomas are non-secretory and therefore do not directly cause hormonal imbalances or systemic side effects (hyperprolactinemia, acromegaly, Cushing’s disease, etc.), they can grow insidiously until they become quite large. Oftentimes, their clinically silent nature means that they will grow until apoplexy occurs as the first sign of their presence. By this time, they are often > 10 mm in diameter, and this explains why patients experiencing pituitary apoplexy most commonly harbor non-functioning pituitary macroadenomas. Nonetheless, although far less common, it is possible for apoplexy to occur in the setting of functioning pituitary macro- and microadenomas. Although these lesions technically possess active secretory endocrine activity, they usually do not manifest with noticeable symptoms or hormonal side effects prior to apoplexy; when apoplexy occurs in the setting of a functioning pituitary adenoma it is often the first presenting sign, just as is the case for nonfunctioning adenomas. According to Nakhleh and colleagues, no differences in endocrine or neuro-opthalmic outcomes were observed between patients with functioning and nonfunctioning adenomas by final follow-up in their retrospective study spanning over two decades [59]. Just as clinical course did not vary by adenoma subtype, consistent results were obtained for patients who underwent conservative and surgical management [59].

Limitations

We acknowledge several limitations to the present study. First and foremost, its retrospective nature as a study featuring heterogeneous data makes it difficult to rule out potential sources of bias. Therefore, the results would benefit from validation through large, multi-institutional randomized controlled trials or registry database studies to investigate the influence of intervention timing and modality on both visual and non-visual clinical outcomes.

Additionally, the analysis was unable to determine specific recovery rates based on the type of visual deficit due to the heterogeneity in the manner by which visual deficits were reported. Unfortunately, subgroup analyses of visual recovery based on the type of presenting visual deficit could not be conducted. Visual recovery was mainly assessed qualitatively, relying on subjective patient-reported improvement in vision as the primary outcome measure. Future investigations should include objective ophthalmologic testing methods such as optical coherence tomography (OCT) and perimetry to quantify improvements in visual deficits before and after surgical intervention. Positive-publication bias is another potential concern, as published data may be more likely to report positive outcomes rather than negative ones in relation to postoperative vision changes. This bias could artificially inflate the reported visual recovery rates. Future prospective or registry studies should aim to evaluate the effects of timing of surgical decompression on clinical outcomes in patients with pituitary apoplexy. Finally, it is unclear whether the population of patients included in this study are representative of the general population of apoplexy patients. The main factor suggesting this is that males comprised approximately 84% of the study population, even though previous studies have consistently reported that apoplexy demonstrates a male predominance ranging from 1.1 to 2.3/1 [60,61,62]. The patients included in the present study exhibit a fairly extreme male predominance, suggesting that the generalizability of the study population is a potential limitation of our meta-analysis.

In summation, while this meta-analysis provides a foundation for future prospective or registry studies, it is important to acknowledge the limitations of the available literature regarding timing of intervention, intervention modality, and functional recovery in patients with pituitary apoplexy. The analysis supports the benefits of surgical intervention, but further investigation is needed to determine whether there is an optimal timing cutoff that must be adhered to when managing pituitary apoplexy that would prevent potentially devastating ocular consequences.