Traditionally, employer-sponsored insurance has been the backbone of the US health coverage system, restricting healthcare access to those with the ability to work full-time [19]. As a result, older Americans and those with low incomes often did not have health insurance. In 1965, the Medicare and Medicaid programs were established in the US, extending health coverage to elderly, low-income, and disabled Americans [20]. Despite these legislative efforts, 7.9% of the US population remains uninsured [21]. These patients often have limited access to healthcare outside of emergency services [22], leading to more severe disease at the time of presentation as noted across both neurosurgical and non-neurosurgical fields [14, 23,24,25]. Moreover, individuals with Medicaid often experience issues with the consistency and adequacy of care due to variations in state-level Medicaid programs [26].

In this study, we found that uninsured patients presented with more advanced disease, had higher preoperative TV leading to reduced EOR, and increased LOS compared to insured counterparts. Patients who experienced lower preoperative KPS, increased VD, preoperative TV, Knosp 4 grade, STR, and PLOS were more likely to be uninsured.

Upon subgroup analysis, Medicaid patients had increased preoperative TV, presented in advanced stage with associated comorbidities, and experienced longer LOS compared to non-Medicaid patients. Those who experienced Knosp 3 grade, STR, and PLOS were more likely to have Medicaid insurance compared to those with private insurance.

Uninsured patients had a significantly lower preoperative KPS (85 [80–90] vs. 90 [80–90]; p = 0004). While Medicaid patients had lower preoperative KPS compared to Medicare and Private (80 [80–90] vs. 90 [80–90] vs. 90 [80–90]), this difference was not statistically significant. To our knowledge, there is no other literature analyzing differences in preoperative KPS by insurance status or type.

VD is one of the most prominent presentations of NFPA, with a recent meta-analysis on 35 case series reporting the incidence ranging from 28 to 100% [27]. In our study, uninsured patients were more likely to present with VD (82.50% vs. 55.81%; p = 0.002). On subgroup analysis of insured patients, Medicaid patients were more likely to present with VD than Private and Medicare patients (73.91% vs. 53.74% vs. 57.85%, respectively), although the difference did not reach statistical significance. This may be due to limited access to early care, leading to delayed presentation with more advanced symptoms. Similar to our findings, Osorio et al. noted a higher prevalence of VD among Medicaid patients compared to Private and Medicare (62.5% vs. 46.7% vs. 44.2%, respectively). However, they also failed to achieve statistical significance [10]. Contrary to Osorio et al. and our findings, Younus et al., including both NFPA and FPA in their cohorts, found no differences in rates of VD among Medicaid and non-Medicaid patients (44% vs. 43%, respectively) [13].

We also noted an increasing trend of apoplexy in uninsured patients (10.00% vs. 6.20%; p = 0.317). Jahangiri et al. previously reported that lack of insurance was significantly associated with pituitary apoplexy [28]. In their study, 11.85% of patients were uninsured as compared to 6.13% in our study. We believe that since our study had a smaller proportion of uninsured patients, the difference in rates of apoplexy did not reach statistical significance. Moreover, upon subgroup analysis, our results were similar to Younus et al. who did not show a statistically significant difference in rates of apoplexy among Medicaid and non-Medicaid groups [13]. To our knowledge, there are no other studies on rates of apoplexy within patients with different types of insurance providers.

Our study used the ellipsoid formula to calculate tumor volume based on the maximum dimensions in axial, coronal, and sagittal planes. Although we recognize that volume segmentation software could yield more precise measurements, we lacked segmentation-based volume data in our dataset. However, as a natural sequela to delayed presentation, we noted that uninsured patients were more likely to have larger TV(13.68 ± 16.89 vs. 6.62 ± 10.27 cm³, p = 0.005) with significantly higher rates of chiasm compression and Knosp 4 grade. Similarly, on subgroup analysis, Medicaid patients had larger TV compared to Private and Medicare groups (14.28 ± 28.82 vs. 6.25 ± 9.56 vs. 6.43 ± 7.19 cm³, respectively), higher rates of chiasm compression, and Knosp 3 grade. There is a lack of literature regarding tumor characteristics among insured and uninsured patients. However, literature combining both FPAs and NFPAs noted that Medicaid patients had larger mean tumor diameters (26.1 mm vs. 23.1 mm) [13]. Regarding the severity of cavernous sinus invasion, in studying patients with NFPA, Osorio et al. noted higher rates of cavernous sinus invasion in Medicaid patients compared to Private and Medicare patients (62.5% vs. 33.7% vs. 43.0%, respectively; p = 0.009) [10]. Younus et al., with both FPAs and NFPAs in their study, failed to show a significant difference in cavernous sinus invasion between Medicaid and non-Medicaid groups [13]. This may be since 27% of patients in their study had FPAs which typically present as microadenomas. Keeping in mind a residual tumor doubling time of 3.4 years [29], it could be extrapolated that patients lacking insurance and Medicaid roughly present with a 3-4-year delay compared to their insured counterparts.

The lower EOR in uninsured patients (85.44 ± 26.03% vs. 96.46 ± 9.60%) is likely related to greater Knosp 4 grade in those patients. This aligns with previous studies associating higher grades of cavernous sinus extension with lower EOR [30, 31]. Upon subgroup analysis, our results correlate with other studies that do not find a significant difference between EOR among patients with different insurance providers [10, 13].

Our study uniquely demonstrates that uninsured patients have a median LOS that is 2 days longer compared to insured patients (p < 0.001). Similarly, the median was 2 days longer in Medicaid patients compared to Medicare and Private patients (p < 0.001). A previous study comparing NFPA noted that private insurance patients were discharged 1 day earlier than those with Medicare and Medicaid, although this did not reach statistical significance. With both FPAs and NFPAs in their cohort, Younus et al. noted that Medicaid patients had significantly longer LOS (9.4 vs. 3.55 days) compared to non-Medicaid patients [13]. Moreover, privately insured patients experienced shorter LOS [32]. The rate of overall complications and hyponatremia was higher in uninsured patients (67.50% vs. 57.93% and 17.50% vs. 11.46%, respectively), and this may be a factor responsible for PLOS. Among insured patients, the longer LOS in Medicaid patients could be attributable to a higher incidence of hyponatremia compared to Private and Medicare patients (21.74% vs. 11.31% vs. 10.31%). Similarly, a higher prevalence of diabetes mellitus in Medicaid patients could be a contributing factor as well [33].

While the rates of overall complications, postoperative CSF leak, transient DI and hyponatremia were higher in uninsured group, the differences were not statistically significant. Notably, despite being significantly younger than the insured group, the uninsured group showed a trend toward higher complication rates. Further analysis using propensity score matching in future studies may help uncover statistically significant differences in complication rates related to insurance status. Subgroup analysis revealed an increased rate of overall complications and transient DI in Private group. However, when transient DI was excluded from the overall complication rate, the difference between groups was not significant anymore. Moreover, there were no differences in other specific complications between patients with different insurance providers which aligns with a previous study by Osorio et al. [10]. Younus et al., with FPAs and NFPAs into their cohort, found a statistically significant increase in postoperative complications in Medicaid group compared to non-Medicaid (14% vs. 7%; p < 0.05) [13]. Performing propensity score matching to further evaluate the impact of insurance on complication rate may lead to emergence of statistically significance. On the other hand, high volume facilities and surgeons with high surgical caseload have been shown to have better short-term outcomes after ETPS [34, 35], potentially equalizing our complication rates across insurance groups despite advanced disease in uninsured and Medicaid patients.

Lack of insurance was associated with reduced preoperative KPS, higher VD, preoperative TV, Knosp 4 grade, STR, and PLOS. Similarly, Medicaid was associated with Knosp 3 grade, STR, and PLOS (p < 0.05). These results are consistent with other studies in neurosurgical literature showing uninsured patients and those with Medicaid to have more severe disease upon presentation and worse postoperative outcomes [6, 12, 28, 36]. This delay in presentation may be due to reduced access to primary care physician or specialty consultation with lack of insurance and underinsurance [36, 37].

Our study is inherently limited by its retrospective nature. To overcome these limitations, we included only patients with complete records available. Additionally, it is a single-institution study, and our results may be influenced by the social and geopolitical setting of our institution. Therefore, larger multi-institutional studies and creation of national pituitary adenoma dataset with granular data is warranted to further investigate our findings. In our study, the probability of cavernous sinus invasion was determined using the Knosp-Steiner classification, which does not distinguish between grade 3 A and grade 3B. Future studies may consider the modified Knosp classification to enable further subgroup analysis based on these distinctions. Another limitation of our study is the absence of data on postoperative hormonal status and long-term hormone replacement needs. Future studies could help clarify the impact of insurance status on postoperative pituitary function and endocrine recovery.

Despite these limitations, our study provides valuable insights into the relationship between insurance and healthcare outcomes in patient undergoing ETPS for NFPA. Unlike existing generic database studies, we were able to collect detailed information on preoperative characteristics, tumor features, and postoperative outcomes, which are not always available in broader datasets. This granularity allows us to highlight significant healthcare disparities driven by insurance status, revealing how these factors contribute to delayed presentation, advanced disease, and prolonged length of stay.