The sellar region is a relatively common site for brain tumors[12, 13]. Over the past 30 years, significant advances in neurosurgery, neuroimaging, and molecular biology have changed the evaluation and management of sellar tumors. Nevertheless, changes in hypothalamus-pituitary function and CNS infection are still frequent postoperative complications and reasons for hospital readmissions[14–17]. The incidence of hypothalamus-pituitary dysfunction is reported to be 4.2/100,000 per year, without sex differences; the prevalence is 45.5 per 100,000 people. The rate of PCNSIs in patients with sellar region tumors ranges from 0.5–14%. Some studies have shown the relationship between abnormal immune function and hypothalamus-pituitary dysfunction. Overall, it is generally recognized that patients with CNS infection might experience hypothalamus-pituitary hormone dysfunction.
In this study, we investigated pituitary hormone levels in patients with sellar region tumors and screened out those with hypothalamus-pituitary dysfunction before PCNSI. Ultimately, we found postoperative DI and postoperative adrenal insufficiency to be independent factors influencing PCNSI. Thus, we suggest that perioperative hypothalamus-pituitary dysfunction may be an underlying cause of PCNSI.
As our multivariate logistic regression analysis indicated, postoperative adrenal insufficiency significantly affected the occurrence of PCNSI[20, 21]. According to a previous study, much of the excess mortality in patients with adrenal insufficiency is attributable to infectious diseases. Indeed, a functional hypothalamic-pituitary-adrenal (HPA) axis is essential for normal health and life expectancy. Furthermore, central adrenal insufficiency is a life-threatening disorder associated with increased morbidity and mortality. The possible infection-related pathogenesis pathways may involve dysregulated systemic inflammation resulting from inadequate intracellular glucocorticoid-mediated anti-inflammatory activity.
In addition, postoperative DI is a risk factor for PCNSI, and we speculate that the reason is impairment of plasma sodium homeostasis. Postoperative hyponatremia and hypernatremia in neurosurgical patients are typically caused by the development of DI. Based on inconsistencies in the definition of DI across the literature, the reported incidence of postsurgical central DI varies from 1 to 67%[25–28]. The course of postoperative DI may be transient, persistent, or triphasic. In the typical triphasic response, a polyuric phase of DI is followed by an oliguric phase of SIADH and then by a third and final phase of persistent DI. When water deficits occur due to inadequate water intake to compensate for polyuria, symptoms of dehydration and/or hyperosmolality develop. Patients may present with hypernatremia in the first and third phases of DI[16, 29, 30]. In addition, a hyponatremic state may result in severe metabolic derangement, myocardial depression and injury, neurologic impairment, venous thromboembolism, and poor wound healing. Moreover, patients with dehydration and hyperosmolality might experience a range of neurological symptoms, including irritability, cognitive decline, disorientation, and confusion, with decreased levels of consciousness, seizure and coma. Various focal neurological deficits may also develop in this context. In general, deterioration of the patient's basic conditions may explain the increased risk of PCNSI that we observed.
Hyponatremia is another common clinical manifestation of DI. Some evidence suggests that sodium is a significant promoter of immune function. Sodium acts by enhancing the function of macrophages and T lymphocytes[33, 34]. A hypernatremic environment may serve as an immunological defense mechanism in inflammatory states, and sodium levels can act in concert with tissue infection as a danger signal, enhancing proinflammatory macrophage and T cell function while dampening anti-inflammatory immune responses. Thus, patients with hyponatremia may show decreased immune function, which may provide an explanation for the frequency of infection among patients with DI. Nevertheless, these studies generally focused on the skin and kidney, and the regulatory circuits that drive salt accumulation in the infected brain are unknown.
PCNSIs occurred more frequently in the TCS cohort than in the TSS cohort in our study. This result is consistent with some previous studies[3, 35]. Currently, transcranial procedures are only applied in special situations, such as for a dumb bell-shaped tumor or one with irregular extensions into the frontal or temporal lobes or when TSS has failed to achieve complete tumor resection. This may result in an increased operation time, an increased risk of postoperative swelling or bleeding of the residual mass and the need for nonbiodegradable materials left at the completion of the TCS. These conditions may lead to an increase in the infection rate. Similarly, patients with a previous history of surgery at the same site are more likely to have central nervous system infections. The reason may be due to the changes in the sellar area structure, which leads to an increase in the difficulty of the operation.
Multivariate logistic regression analysis showed a very strong association of PCNSI development for intraoperative CSF leakage, which is already known. In fact, the risk of postoperative CSF leakage is a major impediment to the use of TSS for resection of sellar lesions, and there are clear correlations between the cranial cavity and the external environment in the event of CSF leakage. Thus, bacteria may more easily enter the cranial cavity from the external environment through the gap and cause a postoperative infection. However, skull base closure techniques have recently evolved, such that CSF leakage is no longer a significant issue following TSS.
As mentioned above, there are many causes associated with PCNSI, yet most of these factors are difficult to correct. A second operation may even be required. In contrast, hypothalamus-pituitary dysfunction can be easily identified and rectified. We believe that the results presented here will help physicians reduce the rate of PCNSI in patients with sellar region tumors.
Because of the retrospective nature of our study, our findings depend on the accuracy of the data recorded in clinical charts, which might have resulted in selection bias. It is hoped that the study findings will prompt future research.