Key Risk Factors for Postoperative Hypotension in Nasopharyngeal Carcinoma Patients After Endoscopic Surgery: A Retrospective Study

Introduction

Nasopharyngeal carcinoma (NPC) is a malignant tumor arising from the epithelial cells of the nasopharyngeal mucosa [1]. It is particularly prevalent in East and Southeast Asia, where radiotherapy remains the mainstay of treatment, often combined with chemotherapy or targeted therapies to improve survival outcomes [2]. However, radiation therapy can lead to severe long-term complications, including progressive vascular, neurological, and glandular damage, which can predispose patients to hemodynamic instability in the perioperative period [3,4]. Among these, xerostomia is the most prevalent radiation-related complication, affecting 91.8–97.5% of patients [5,6]. Additionally, vascular endothelial damage, fibrosis, and autonomic dysfunction resulting from radiation exposure can contribute to altered cardiovascular responses in surgical patients, increasing the risk of perioperative complications [7]. With the advancement of endoscopic surgery, the role of surgical intervention in NPC management has expanded, particularly for recurrent tumors and radiation-induced complications [8]. This minimally invasive approach is associated with fewer surgical complications, improved functional outcomes, and better quality of life for patients [8–11]. However, despite these benefits, postoperative hypotension has emerged as a significant concern, particularly in patients with prior radiation exposure. Clinical data from Guangdong Provincial People’s Hospital indicate that approximately 30% of NPC patients undergoing endoscopic surgery experience postoperative hypotension, often requiring volume resuscitation, blood transfusions, or vasoactive agents such as dopamine, norepinephrine, and epinephrine. Postoperative hypotension is a well-recognized predictor of adverse surgical outcomes, yet its impact in NPC patients remains poorly understood. In surgical populations, hypotension is associated with inadequate tissue perfusion, which can lead to myocardial ischemia, acute kidney injury, postoperative delirium, and increased perioperative mortality [12–14]. Radiation-induced vascular damage can further exacerbate these risks, as prior studies have demonstrated that radiation exposure contributes to arterial stiffness, endothelial dysfunction, and impaired autonomic regulation, all of which can affect blood pressure homeostasis. However, postoperative hypotension following endoscopic surgery in NPC patients remains significantly underreported, and its underlying mechanisms are not well characterized.

Postoperative hypotension is a common yet under-recognized complication following surgery, with significant implications for patient recovery and overall prognosis. Studies have consistently demonstrated that postoperative hypotension is associated with an increased risk of adverse events, including surgical site infections, myocardial injury, and end-organ dysfunction. Yilmaz et al found that postoperative hypotension following colorectal surgery was significantly correlated with an increased incidence of surgical site infections, highlighting the importance of maintaining hemodynamic stability to optimize wound healing and immune function [15]. Similarly, Liem et al reported that hypotension occurring after noncardiac surgery was independently associated with myocardial injury, reinforcing that even brief episodes of hypotension can have lasting cardiovascular consequences [16]. The physiological mechanisms underlying these associations involve inadequate tissue perfusion, reduced oxygen delivery, and systemic inflammatory responses that contribute to poor postoperative outcomes [15,16].

Furthermore, Smischney et al demonstrated that patients who experienced postoperative hypotension requiring ICU admission were at significantly higher risk of adverse clinical outcomes, including prolonged hospitalization and increased mortality [17]. In vascular surgery, Rubio et al identified several predictors of postoperative hypotension following carotid artery stenting, including preoperative blood pressure levels and autonomic dysfunction, which may also be relevant to head and neck cancer patients undergoing surgery after radiation therapy [18]. Namekawa et al investigated postoperative hypotension following laparoscopic adrenalectomy for pheochromocytoma and found that prolonged hypotensive episodes were strongly associated with preoperative cardiovascular instability and intraoperative hemodynamic fluctuations, emphasizing the need for careful perioperative management in high-risk populations [19]. Given the growing body of evidence linking postoperative hypotension to adverse surgical outcomes, enhanced intraoperative blood pressure monitoring and individualized anesthetic management strategies may be crucial in mitigating its impact in patients with pre-existing cardiovascular and autonomic dysfunction, such as those undergoing endoscopic surgery for NPC.

The existing literature does not address whether specific NPC-related factors, such as prior radiation exposure, tumor location, or preoperative cardiovascular status, contribute to an increased risk of postoperative hypotension. Additionally, it remains unclear whether preoperative hemodynamic optimization strategies could reduce this risk and improve postoperative outcomes. This study aimed to determine the incidence and identify risk factors for postoperative hypotension in NPC patients undergoing endoscopic surgery, with a focus on how prior radiation exposure and other patient-specific factors contribute to hemodynamic instability. Understanding these risk factors is crucial for developing targeted perioperative management strategies, optimizing fluid resuscitation protocols, and improving patient outcomes. Findings from this study may inform future research on individualized perioperative care and contribute to refining surgical guidelines for NPC management.

Material and Methods

STUDY DESIGN AND PATIENT SELECTION:

This retrospective study included 264 patients with nasopharyngeal carcinoma (NPC) who underwent endoscopic surgery following radiotherapy at Guangdong Provincial People’s Hospital (Guangzhou, China) between January 2021 and December 2023. Data were extracted from the electronic medical record system of the hospital, including perioperative and postoperative monitoring datasets. To minimize selection bias, all consecutive patients meeting the inclusion criteria were included. All surgeries were performed by a specialized team of ear, nose, and throat (ENT) surgeons. The study was approved by the Ethics Committee of Guangdong Provincial People’s Hospital. Due to its retrospective nature, the requirement for informed consent was waived. To reduce potential biases in retrospective data collection, a standardized protocol was implemented for data extraction and verification. Data inconsistencies were resolved by cross-referencing patient records with surgical and anesthesia documentation.

INCLUSION CRITERIA:

We included patients who underwent endoscopic surgery for NPC recurrence or radiation-induced complications after radiotherapy, aged 18–75 years, classified as American Society of Anesthesiologists (ASA) status I–III, and scheduled for elective surgery. Additionally, patients were also required to have been hospitalized for at least 2 nights postoperatively to ensure adequate hemodynamic monitoring.

EXCLUSION CRITERIA:

Patients were excluded if they exhibited preoperative hypotension, defined as a systolic blood pressure (SBP) below 90 mmHg or mean arterial pressure (MAP) below 65 mmHg to avoid confounding effects on postoperative outcomes. Those who failed to undergo electrocardiogram (ECG) monitoring for at least 4 hours immediately after surgery were also excluded, as were patients with unrecorded electrocardiogram data exceeding 30% of the total monitoring period. Patients who required postoperative admission to the intensive care unit (ICU) were also excluded, as their hemodynamic responses would be significantly influenced by intensive therapeutic interventions, thus introducing confounding variables.

DATA COLLECTION:

All data were retrieved from electronic medical records. Collected variables included demographic information (sex, age, height, weight, body mass index), ASA status, comorbidities (hypertension, diabetes, respiratory disease, cerebrovascular disease), and surgical history. Routinely collected preoperative blood sampling data included hematocrit (HCT), hemoglobin (Hgb), white blood cells (WBC), sodium (Na), potassium (K), calcium (Ca), magnesium (Mg), albumin (Alb), creatinine (Cr), C-reactive protein (CRP), and blood glucose (Glu). Operative data included the duration of anesthesia, duration of surgery, duration of stay in the postanesthesia care unit (PACU), use of dexmedetomidine, total fluid intake, intraoperative bleeding, intraoperative blood transfusion, and intraoperative urine output. Postoperative monitoring included continuous ECG and non-invasive blood pressure measurements (NIBP) recorded every hour for 24 hours using a using a standardized automated blood pressure monitoring system. To ensure data accuracy and consistency, all extracted variables were cross-validated against surgical, anesthesia, and nursing documentation. Missing values were addressed using multiple imputation techniques where appropriate, and outliers in continuous variables were assessed using Tukey’s method. Patients were classified into 2 groups based on their postoperative blood pressure readings: (1) the Postoperative Hypotension Group (PHG) consisted of patients with SBP <90 mmHg or MAP <65 mmHg (meeting either criterion), and the No Hypotension Group (NHG) was composed of patients who did not meet the criteria for hypotension.

STATISTICAL ANALYSIS:

Continuous variables are expressed as means±standard deviations (SDs), while categorical variables are expressed as numbers (percentages). Initially, univariate analyses were conducted to assess crude associations between variables. Variance inflation factors (VIFs) were calculated to evaluate collinearity, and variables with severe collinearity (VIF >10) were excluded from the multivariate analysis. Variables with a P value <0.1 from the univariate analysis were included in the multivariate logistic regression model. Multivariate logistic regression was performed using a forward selection approach to identify predictors of postoperative hypotension. Assumptions for logistic regression, including linearity in the logit, absence of multicollinearity, and adequate model fit, were assessed before finalizing the model. Statistical significance was set at P<0.05. All statistical analyses were conducted using SPSS version 25.0 (SPSS, Inc., Chicago, IL, USA). To ensure reproducibility, the choice of statistical techniques was based on best practices for retrospective cohort studies, and model assumptions were carefully validated before interpretation of results.

Results

PATIENT SELECTION AND BASELINE CHARACTERISTICS:

This study included 326 patients with nasopharyngeal carcinoma (NPC) who underwent endoscopic surgery after radiotherapy. Following application of the inclusion and exclusion criteria, 13 patients were excluded due to postoperative admission to the intensive care unit (ICU), while 49 patients were excluded due to missing postoperative electrocardiogram (ECG) data. Ultimately, a total of 264 patients were included in the final analysis, as illustrated in Figure 1. To ensure methodological consistency, missing values were handled using multiple imputation, and patient selection was conducted using predefined criteria to reduce selection bias. The baseline characteristics of the study population are summarized in Tables 1 and 2. The cohort consisted of 185 male and 79 female patients, with an average body mass index (BMI) of 21.15 kg/m2 and a mean age of 50 years. The postoperative hypotension group (PHG) comprised 86 patients (32.5%), while the no hypotension group (NHG) included 178 patients (67.4%).

COMPARISON OF BASELINE CHARACTERISTICS BETWEEN PHG AND NHG:

The proportion of females was significantly higher in the PHG (44.2%) than the NHG (23.0%) (P=0.001). Patients in the PHG had a shorter average height (162.5±8.0 cm) than those in the NHG (165.6±7.9 cm; P=0.011). Similarly, mean weight in the PHG was lower (55.2±13.7 kg) compared to the NHG (58.9±14.2 kg; P=0.044). PHG patients had a significantly lower preoperative SBP (112.5±23.4 mmHg) than NHG patients (120.4±14.8 mmHg; P=0.001). A similar trend was observed for mean arterial pressure (MAP), with the PHG having a lower MAP than the NHG (P=0.007). Preoperative blood test results also revealed significant differences, including lower hemoglobin levels (106.87±19.65 g/L) in the PHG compared to the NHG (115.62±18.43 g/L; P<0.001), lower blood calcium levels (2.23±0.10 mmol/L vs 2.27±0.09 mmol/L, P=0.006), and slightly lower albumin levels (35.53±3.22 g/L vs 36.70±5.04 g/L, P=0.049).

INTRAOPERATIVE AND POSTANAESTHETIC CARE UNIT (PACU) DATA:

Intraoperative parameters and PACU data are presented in Table 2. Dexmedetomidine use was higher in PHG (91.8%) than NHG (83.7%), but the difference was not statistically significant (P=0.076). Other parameters, including operative duration, fluid intake, urine volume, blood loss, and time in the PACU, did not differ significantly between the groups. A more detailed exploratory analysis was conducted to evaluate potential confounders, but none were found to significantly alter the association between dexmedetomidine use and postoperative hypotension.

PREDICTORS OF POSTOPERATIVE HYPOTENSION:

Univariate and multivariate logistic regression analyses were performed to identify independent predictors of postoperative hypotension, adjusting for confounders identified through stepwise selection. The results are presented in Table 3. Height (OR=0.951, 95% CI 0.917–0.986, P=0.006), admission SBP (OR=0.979, 95% CI 0.963–0.995, P=0.010), preoperative blood calcium levels (OR=0.036, 95% CI 0.002–0.708, P=0.029), and intraoperative use of dexmedetomidine (OR=0.278, 95% CI 0.091–0.848, P=0.001) were negatively associated with postoperative hypotension. While the association between height and reduced risk of postoperative hypotension has been observed in many studies [20], its relevance in NPC patients remains underexplored. Lower preoperative calcium levels may contribute to altered vascular tone, impairing vasoconstriction, while protective effect of dexmedetomidine may be attributed to its ability to maintain hemodynamic stability through selective alpha-2 receptor agonism. These findings suggest a potential interplay between patient-specific physiological characteristics and perioperative pharmacologic interventions.

RECEIVER OPERATING CHARACTERISTIC (ROC) ANALYSIS:

The ROC curve for the predictive model, derived from multivariate logistic regression, had an area under the curve (AUC) of 0.703 and a sensitivity of 0.782 (Figure 2). These findings suggest that the model has moderate predictive accuracy for identifying patients at high risk for postoperative hypotension. Although further external validation is required, these results provide an initial framework for early risk stratification in NPC patients undergoing endoscopic surgery.

POSTOPERATIVE OUTCOMES:

There were 3 cases of poor postoperative outcomes, all of which occurred in the PHG. One patient experienced a large cerebral infarction and was transferred to Neurosurgery for further treatment. Another developed aspiration pneumonia and, due to a poor prognosis, was discharged against medical advice. A third patient died due to complications related to severe bleeding after nosebleeds that could not be controlled.

Discussion

Postoperative hypotension lacks a unified clinical standard, leading to widely varying reported incidence rates, from 5% to 93%, depending on the criteria used [21]. Sessler et al [12] found that perioperative interventions targeting systolic blood pressure (SBP) <90 mmHg are associated with an increased risk of myocardial infarction during surgery and hospitalization. Similarly, Wesselink et al [14] reported that maintaining mean arterial pressure (MAP) ≤65–60 mmHg for 5 minutes or longer moderately increases the risk of terminal organ damage. Given these findings, this study defined postoperative hypotension as SBP <90 mmHg or MAP <65 mmHg, revealing an incidence rate of 32.5% in NPC patients undergoing endoscopic surgery. This aligns with previous studies on head and neck surgeries, such as Cirbian et al [22], who reported a 34% incidence of postoperative hypotension following carotid stent placement using an SBP threshold of <90 mmHg. While the hourly blood pressure monitoring approach in this study improves feasibility, it may underestimate transient episodes of hypotension. Liem et al [23] reported that 8% of patients undergoing noncardiac surgery experienced MAP <60 mmHg within 2 hours postoperatively when continuous monitoring was used, indicating that short-lived hypotensive episodes may not have been captured. Future studies employing continuous monitoring could provide a more comprehensive assessment of hypotension incidence in NPC patients. Several factors contribute to postoperative hypotension, including patient-related characteristics, surgical factors, and anesthesia-related complications [24]. In this study, independent predictors of postoperative hypotension included height, admission SBP, preoperative blood calcium levels, and intraoperative use of dexmedetomidine. These findings suggest that radiation-induced vascular injury, electrolyte disturbances, and anesthetic choices may play a key role in postoperative hemodynamic stability.

An unexpected finding was that height emerged as a protective factor for postoperative hypotension (OR=0.951, 95% CI 0.917–0.986, P=0.006), contradicting previous studies that have suggested an inverse relationship between height and blood pressure regulation [25–27]. Song et al [26] and Cossio et al [28] proposed that shorter individuals may have earlier return of arterial reflection waves, leading to increased central pressure during the late contraction phases of the cardiac cycle. This phenomenon could explain why shorter individuals are more prone to postoperative hypotension. Studies on cardiovascular hemodynamics have also demonstrated that arterial compliance varies with body height, influencing perfusion pressures during physiological stress [29,30].

Preoperative admission SBP was also identified as a protective factor, with an optimal cutoff of 100.5 mmHg based on the ROC curve analysis. These findings are consistent with Dogrul et al [31], who found that lower preoperative SBP was an independent risk factor for postoperative complications in elderly surgical patients. One hypothesis for this association is that patients with lower baseline SBP may have reduced intravascular volume or autonomic dysfunction, making them more susceptible to perioperative hemodynamic instability, especially after radiation exposure [32]. Reduced vascular tone and autonomic impairment have been implicated in post-radiotherapy cardiovascular dysfunction, supporting this finding [33].

This study identified lower preoperative blood calcium levels as an independent predictor of postoperative hypotension, consistent with studies linking hypocalcemia to impaired vascular tone and autonomic instability [34]. Nearly 53% of NPC patients experience hypocalcemia following radiotherapy, likely due to radiation-induced parathyroid dysfunction, which can contribute to altered vasoregulation [35]. The identified calcium threshold of 2.275 mmol/L for predicting postoperative hypotension suggests that subclinical hypocalcemia could serve as a predisposing factor in this patient population. Newer research has also linked calcium metabolism disruptions to increased risks of perioperative hemodynamic instability, further supporting the need for targeted electrolyte management in these patients [36]. The intraoperative use of dexmedetomidine was associated with an increased risk of postoperative hypotension. Dexmedetomidine, a selective α-2 adrenergic agonist, provides sedation and analgesia with opioid-sparing benefits, but its vasodilatory effects can lead to intraoperative hypotension and bradycardia [37–39]. A growing body of literature indicates that dexmedetomidine can impair compensatory vasoconstriction mechanisms, particularly in patients with pre-existing endothelial dysfunction or radiation-induced vascular damage [37–43]. Despite its widespread use in head and neck surgery due to its ability to improve intraoperative visualization and reduce sympathetic activation, its hemodynamic effects require careful consideration, particularly in NPC patients with pre-existing vascular damage. These findings align with previous studies indicating that dexmedetomidine increases the risk of intraoperative hypotension [44,45], necessitating individualized dosing strategies to optimize hemodynamic stability. Given that dexmedetomidine has well-documented benefits in maintaining stable intraoperative conditions, alternative intraoperative management strategies should be explored. For instance, individualized dosing regimens, combining dexmedetomidine with lower doses of vasopressors, or adjusting its infusion rate in patients with pre-existing cardiovascular risks, may mitigate its hypotensive effects without compromising its sedative advantages. Recent anesthetic protocols have suggested modified dexmedetomidine infusion strategies that maintain its benefits while reducing hypotensive risk in high-risk patients [46].

This study has several limitations that should be acknowledged. First, while multiple independent predictors of postoperative hypotension were identified, other potential influencing factors such as medication history, anesthesia mode, intraoperative blood pressure variability, and postoperative fluid resuscitation strategies were not fully analyzed. Future studies should incorporate a broader range of perioperative variables to gain a more comprehensive understanding of hypotensive risk factors in NPC patients. Second, while all patients received radiotherapy prior to surgery, the specific radiation doses and regimens were not documented. Given that radiation therapy can influence vascular compliance and autonomic function, variations in radiation exposure could contribute to differences in postoperative hemodynamic responses. Future prospective studies should incorporate detailed radiation dose data to clarify these effects. Third, intraoperative blood pressure variations were not systematically recorded in this study. While hourly postoperative blood pressure monitoring provided valuable insight into persistent hypotension, continuous intraoperative monitoring may have captured transient episodes that could contribute to adverse postoperative outcomes. Future studies should integrate continuous hemodynamic monitoring to provide a more comprehensive assessment of intraoperative and postoperative blood pressure fluctuations. Finally, this was a single-center retrospective study, which may limit the generalizability of the findings. Although statistical adjustments were applied to account for confounders, prospective multicenter trials are needed to validate these results in a larger and more diverse patient population.

Conclusions

This study found that 32.5% of nasopharyngeal carcinoma (NPC) patients undergoing endoscopic surgery after radiotherapy experienced postoperative hypotension. Height, admission systolic blood pressure (SBP), and preoperative calcium levels were protective factors, while intraoperative dexmedetomidine use increased hypotension risk. These findings highlight the need to identify high-risk patients, optimize perioperative management, and carefully regulate dexmedetomidine use to improve outcomes. Further research is needed to clarify the physiological basis of height as a protective factor, particularly in relation to vascular compliance and autonomic regulation. Additionally, radiation dose variability was not accounted for in this study, warranting future investigations into its role in blood pressure regulation. Prospective trials should also refine dexmedetomidine dosing strategies to balance its benefits and hemodynamic risks. By improving risk stratification and perioperative management, future research may help enhance patient safety and surgical outcomes in NPC populations.