Complications of controlled hypotension during functional endoscopic sinus surgery – a systematic review

Abstract

Background

Although functional endoscopic sinus surgery (FESS) constitutes a standard treatment for the management of chronic rhinosinusitis, intraoperative bleeding poses a considerable challenge by impairing visibility and increasing complication risks. Controlled hypotension is a widely employed technique to mitigate bleeding; however, excessive blood pressure reduction may compromise systemic perfusion and predispose to organ ischemia.

Objectives

Due to the scarcity of literature evaluating the safety of controlled hypotension in FESS, we aimed to systematically review current evidence on the risks and potential complications of this technique, with an emphasis on systemic adverse events (hemodynamic instability, neurological impairment, and renal dysfunction).

Methods

The PubMed, Embase, and Cochrane databases were searched for English-language original articles published until October 1, 2025. Studies were considered eligible if they reported mean arterial pressure (MAP) values achieved and evaluated systemic complications associated with controlled hypotension during FESS.

Results

A total of 32 articles were found eligible and included in the further analysis. Most protocols aimed for a MAP of 50–70 mmHg, primarily using dexmedetomidine, propofol, and remifentanil. Only 8 of the 32 studies utilized advanced intraoperative monitoring (e.g., near-infrared spectroscopy [NIRS], transcranial Doppler [TCD], and non-invasive cardiac output monitoring [NICOM]). The most frequent minor complication was postoperative nausea and vomiting (in 2–62% of cases). Significant hemodynamic disturbances included intraoperative bradycardia and severe hypotension (MAP < 50 mmHg), requiring pharmacological intervention. While no permanent neurological deficits were reported, intraoperative cerebral desaturation was observed in 2 studies. The follow-up observation rarely extended beyond the post-anesthesia care unit and did not include a structured assessment of cerebral and renal function.

Conclusions

Subclinical complications of controlled hypotension during FESS may be overlooked without advanced intraoperative monitoring, postoperative biomarker surveillance, and a long-term follow-up period. The application of a uniform MAP target of 50–70 mmHg may be insufficient in high-risk patients. The development of personalized hemodynamic management strategies and enhanced perfusion-monitoring protocols is required to reduce the risk of occult organ ischemia and to define safe limits for controlled hypotension in FESS.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-026-04021-5.

Introduction

Functional endoscopic sinus surgery (FESS) is a widely used technique for the management of chronic rhinosinusitis, nasal polyposis, and other disorders of the nasal cavity and paranasal sinuses. The procedure involves the insertion of an endoscope and specialized micro-instruments into the nasal cavity to restore patency of the natural sinus ostia and remove pathological tissue [1]. FESS is considered minimally invasive, with major complication rates ranging from 0.3% to 3% [2]. Reported adverse events include injury to orbital structures, excessive lacrimation, cerebrospinal fluid leakage, and intracranial complications. Life-threatening events such as massive hemorrhage requiring transfusion or intraoperative mortality remain exceedingly rare.

Among all complications, intraoperative bleeding is the most frequent, accounting for approximately 23–40% of cases. Even relatively small volumes of blood might significantly impair endoscopic visibility, increasing the risk of iatrogenic injury or the need to discontinue the procedure [2, 3]. Controlled hypotension is therefore a widely applied anesthetic strategy in FESS, intended to reduce bleeding and improve surgical field clarity. While physical positioning (reverse Trendelenburg position or isolated head elevation) may assist in lowering blood pressure, pharmacological agents that reduce systolic blood pressure (SBP)/mean arterial pressure (MAP) – β-blockers, α2-agonists, nitrates, intravenous and inhalational anesthetics, and clonidine – remain the primary method [4, 5].

Although controlled hypotension improves visibility and shortens operative time, excessive blood pressure reduction (SBP to 80–90 mmHg or MAP to 50–70 mmHg) may compromise systemic perfusion and predispose to cerebral, renal, myocardial, splanchnic, or hepatic ischemia (Fig. 1) [3, 4]. Reported complications remain rare (approximately 0,6%), and perioperative mortality is negligible in several studies [6, 7]. Nevertheless, evidence evaluating the microcirculatory and systemic safety of controlled hypotension in FESS is limited, as most research focuses primarily on bleeding reduction and surgical field grading.

Fig. 1.

Hemodynamic impact of controlled hypotension. MAP – Mean Arterial Pressure; SBP – Systolic Blood Pressure; DBP - Diastolic Blood Pressure; AKI – Acute Kidney Injury

The aim of this systematic review is to synthesize current evidence on the risks and potential complications associated with controlled hypotension during FESS, with an emphasis on the systemic adverse events – hemodynamic instability, neurological impairment, and renal dysfunction.

Methods

Study guidance

The review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [8].

Search strategy and criteria

The PubMed, Embase, and Cochrane databases were searched independently by three authors (R.K., W.SG., L.Z.) for English-language papers published from 1st January 2000 to 1st October 2025. The search strategy combined the following terms (“complications” OR “adverse effects” OR “silent myocardial infarction” OR “myocardial injury” OR “cardiac ischemia” OR “troponin” OR “regional cerebral oxygen saturation” OR “stroke” OR “cerebral ischemia” OR “cerebral oximetry” OR “ischemic injury” OR “cerebral perfusion” OR “acute kidney failure” OR “acute renal insufficiency” OR “acute renal injury” OR “creatinine”) AND (“controlled hypotension” OR “deliberate hypotension” OR “hypotensive anesthesia” OR “mean arterial pressure”) AND (“sinus surgery” OR “endoscopic sinus” OR “FESS” OR “ESS”). Generic isolated terms such as “hypotension” or “intraoperative hypotension” were not used as stand-alone intervention terms because the review question focused on intentional controlled hypotension rather than unplanned or incidental hypotensive episodes. Records retrieved under MAP and controlled/deliberate hypotension terminology were therefore screened specifically for intentional hypotensive protocols used during FESS.

After the removal of duplicates, two authors (R.K. and L.Z) independently screened the titles and abstracts of the remaining records to identify studies meeting the following inclusion criteria: (1) study design: clinical studies, clinical trials, and randomized controlled trials (RCTs); (2) population: patients aged 16 years and older; (3) articles published in English; (4) intervention: controlled hypotension during FESS induced by intravenous drug administration, and (5) outcome: reporting complications or adverse events (specifically, cerebral/renal ischemia and cardiovascular instability; with the exclusion of surgical field visibility/bleeding assessments).

Studies were excluded at this stage if they: (1) constituted reviews, meta-analyzes, case reports or case series (n < 5 patients); (2) included patients aged 15 years and younger; (3) were non-English articles; (4) were published before 1st January 2000; (5) described surgical procedures other than FESS (e.g., septoplasty, rhinoplasty, or non-endoscopic sinus surgery – from external approach); (6) did not describe the application of controlled hypotension or used exclusively inhalational or oral hypotensive protocols, and (7) did not report complications or adverse events associated with controlled hypotension.

Additionally, the reference lists in all preselected articles were screened for further relevant studies.

Eligibility criteria

Full-text articles were retrieved and assessed for eligibility by two authors (R.K. and W.GS.). The study was considered eligible if it: (1) utilized an intravenous controlled hypotension protocol during FESS; (2) specified target MAP levels; (3) reported the incidence rates of perioperative complications, and (4) described the intraoperative monitoring techniques and patient follow-up.

Any disagreements during the screening and selection phases were first resolved through discussion until a consensus was reached. If consensus could not be achieved, the final decision was adjudicated by a third reviewer (M.A.).

Data extraction

From the included studies, the following data were extracted: (1) hypotensive agent: a specific intravenous drug used for controlled hypotension; (2) hemodynamics: achieved mean arterial pressure (MAP) values and, where reported, heart rate targets or heart-rate-modifying agents; (3) complications: incidence rates of adverse; (4) intraoperative monitoring modalities (basic and advanced); and (5) duration and details of patient follow-up; and (6) when available, surgical duration and surgical field visibility/bleeding scores. Surgical field quality and operative duration were noted descriptively but were not incorporated into the primary safety synthesis, because the main objective of this review was to evaluate complications and monitoring rather than comparative efficacy.

Assessment of study quality and risk of bias in included studies

Three authors (M.A., R.K., and L.Z) independently assessed the risk of bias for each included article using the revised Cochrane Risk of Bias Tool for Randomized Trials (RoB 2) for RCTs, and the Risk of Bias in Non-randomized Studies of Interventions tool (ROBINS-I) for non-randomized studies. Any discrepancies were resolved through discussion until consensus was reached.

Results

Study selection

The initial database search identified 1228 records, including 286 from PubMed, 515 from Embase, and 427 from Cochrane. After removing 866 duplicates, 362 records were screened. One hundred ninety-two reviews, systematic reviews, or meta-analyses were excluded, and 99 articles on irrelevant topics were rejected, as well as 9 case reports, 6 non-English language, 5 articles with patients aged 15 and younger, and 2 case series (n < 5 patients). After the exclusion of 313 articles that did not meet the predefined criteria, the full text of 49 reports was sought for retrieval, of which 5 could not be obtained.

A total of 44 full-text articles were assessed for eligibility. Following detailed evaluation, 12 were excluded because 1 was published before 1st January 2000, 2 were non-English-language articles, and the remaining 9 were irrelevant to the research question. Additionally, 12 articles were identified through a manual backward citation chaining. Of these, 1 could not be retrieved, and the remaining 11 were excluded because 2 were duplicates, 2 were published before 1st January 2000, and 7 were reviews or systematic reviews.

A total of 32 articles were found eligible and included in the further analysis. Figure 2 shows the entire literature selection process.

Fig. 2.

PRISMA flowchart of the medical database search strategy

Study characteristics

The analysis of 32 included studies (15 evaluating single interventions [9–23] and 17 comparative studies [24–40]) rendered 2422 patients. Among the studies included in the systematic review, 4 were cohort studies [9–12] (evaluating a total of 328 patients), and 28 constituted randomized controlled trials [13–40] (enrolling a total of 2094 individuals). Complete study characteristics are presented in Table 1.

Table 1.

Characteristics of the included studies

Author, year	Study design	Number of patients	Mean age (SD) (years)	Used drugs/ Compared drugs	Achieved MAP range (mmHg)
Cincikas et al.; 2003 [9]	Cohort study	52	44.1 (11.2)	Nitroglycerin	50–60
Naik et al.; 2011 [10]	Cohort study	213	16–67*	Propofol, Remifantanyl	60–70
Ha et al.; 2014 [11]	Cohort study	32	48	Propofol	44–72
Heller et al.; 2015 [12]	Cohort study	31	54.3 (16.1)	Propofol, Remifentanyl, Nitrous Oxide	57–83
Goksu et al.; 2008 [13]	RCT	62	38.46 (10.26)	Dexmedetomidine	85–98
Sartcaoglu et al.; 2005 [14]	RCT	46	35 (10.3)	Nitroglycerine	60–70
Zhang et al.; 2021 [15]	RCT	39	50.55 (14.34)	Nicardipine, Esmolol	67–71
Tirelli et al.; 2004 [16]	RCT	64	48.07 (9.40)	Propofol, Remifentanyl	60–70
Li et al.; 2023 [22]	RCT	110	44.5 (11.8)	Propofol, Remifentanyl	65–75
Jiwanmall et al.; 2017 [17]	RCT	60	18–65*	Clonidine	55–65
Hassan et al.; 2024 [23]	RCT	70	42.26 (11.44)	Dexmedetomidine	55–65
Gollapudy et al.; 2020 [18]	RCT	70	41 (13)	Propofol vs. Desflurane	60–70
Neethirajan et al.; 2020 [19]	RCT	92	35 (13.5)	Dexmedetomidine	-
Rehab et al.; 2023 [20]	RCT	52	28.77 (6.95)	Aminophylline	60–65
Ankichetty et al.; 2011 [21]	RCT	40	31.55 (11.54)	Propofol vs. Isoflurane	60–70
Skalias et al.; 2023 [24]	RCT	72	50.4 (9.5) (B)	Bisoprolol vs. Nifedipine	69–117
			49.1 (12.9) (Nif)
Alkan et al.; 2021 [25]	RCT	60	36.5 (R)	Esmolol vs. Remifentanil vs. Nitroglycerin	60–65
			33.5 (E)
			47.5 (Ntg)
Bharathwaj et al.; 2019 [26]	RCT	80	35.05 (10.91) (D)	Labetalol vs. Dexmedetomidine	60–70
			35.90 (13.41) (P)
Bajwa et al.; 2016 [27]	RCT	150	31.6 (5.2) (D)	Nitroglycerine vs. Esmolol vs. Dexmedetomidine	60–70
			36.4 (6.1) (Ntg)
			34.3 (5.6) (E)
Yun et al.; 2015 [28]	RCT	28	30.5 (Ntp)	Nitroprusside vs. Remifentanil	60–70 (30% decline from baseline)
			30.5 (R)
Sajedi et al.; 2016 [29]	RCT	62	36.36 (12.35) (R)	Remifentanil vs. Labetalol	50–60 (20–30% decline from baseline)
			34.34 (10.91) (L)
Sujatha et al.; 2023 [30]	RCT	100	36.18 (2.54) (D)	Dexmedetomidine vs. Propofol	60–70
			38.30 (7.42) (P)
Praveen et al.; 2024 [31]	RCT	80	45.36 (10.01) (D)	Dexmedetomidine vs. Clonidine	55–65
			44.45 (9.91) (C)
Bafna et al.; 2021 [32]	RCT	70	27.23 (7.75) (D)	Dexmedetomidine vs. Clonidine	65–70
			27.49 (8.68) (C)
Sharma et al.; 2025 [33]	RCT	60	37.43 (5.21) (E)	Esmolol vs. Dexmedetomidine	60–70
			37.3 (7.32) (D)
Escamilla et al.; 2019 [34]	RCT	94	47.60 (11.0) (C)	Clonidine vs. Dexmedetomidine	55–65
			48.02 (10.51) (D)
Eghbal et al.; 2018 [35]	RCT	100	31.7 (9) (L)	Labetalol vs. Dexmedetomidine	65–75
			5.06 (9.64) (D)
Moshiri et al.; 2017 [36]	RCT	100	35.16 (8.2) (D)	Propofol vs. Dexmedetomidine	65–75
			36.90 (8.2) (P)
Huh et al.; 2020 [37]	RCT	140	46 (17) (D)	Dexmedetomidine vs. Remifentanil	65–75
			42 (17) (R)
Karabayirli et al.; 2016 [38]	RCT	47	36 (R)	Dexmedetomidine vs. Remifentanil	55–65
			37 (D)
Mugabo et al.; 2024 [39]	RCT	80	33.90 (6.85) (C)	Dexmedetomidine vs. Clonidine	55–65
			35.15 (6.62) (D)
Lee et al.; 2013 [40]	RCT	66	39.8 (17.4) (R)	Dexmedetomidine vs. Remifentanil	65–75
			45.5 (12.1) (D)

^*, age range, no data for the mean age was provided in the study

RCT Randomized Controlled Trial, MAP Mean Arterial Pressure, TIVA Total Intravenous Anesthesia, B Bisoprolol, Nif Nifedipine, R Remifentanil, E Esmolol, Ntg Nitroglycerin, D Dexmedetomidine, Ntp Nitroprusside, L Labetalol, C Clonidine, P Propofol

Risk of bias analysis

Fourteen of the 28 RCTs [13–15, 17, 21, 24, 26–31, 34, 39], assessed using RoB 2, were judged to have “some concerns” regarding the risk of bias, primarily related to outcome measurement and the selection of the reported results. One trial [16] was rated as posing a high risk of bias arising from the randomization process. Detailed results of the RoB2 assessment are presented in Figs. 3 and 4.

Fig. 3.

Results of the bias risk assessment using RoB 2 for randomized studies. RoB 2, revised cochrane risk of bias tool for randomized trials

Fig. 4.

Proportion of bias risk assessment results using RoB 2 for randomized studies. RoB 2, revised cochrane risk of bias tool for randomized trials

Among the non-randomized studies evaluated with the ROBINS-I tool, two studies [9, 10] were found to pose a high risk of bias, primarily due to confounding and outcome measurement. The detailed results of ROBINS-I analysis are presented in Figs. 5 and 6.

Fig. 5.

Results of the bias risk assessment using ROBINS-I for non-randomized studies. ROBINS-I, cochrane risk of bias tool for non-randomized studies - of interventions

Fig. 6.

Proportion of bias risk assessment results using ROBINS-I for non-randomized studies. ROBINS-I, cochrane risk of bias tool for non-randomized studies - of interventions

Hemodynamic impact and target MAP

The analysis of 32 included studies demonstrated that most protocols successfully achieved the intended MAP of 50–70 mmHg, generally accepted as the threshold balancing favorable operating conditions with maintained perfusion. In two studies [28, 29], the target was defined as a 20–30% reduction in MAP relative to baseline, yet the achieved absolute values remained within the same range. Only one study [19] reported a lack of complete numerical data on the course of MAP during surgery (Table 1).

In the analyzed material, the most frequently studied pharmacological agents used to induce and maintain controlled hypotension were: dexmedetomidine (16/32 studies [13, 19, 23, 26, 27, 30–40]), propofol (9/32 [10–12, 16, 18, 21, 22, 30, 36]), and remifentanil (9/32 [12, 16, 22, 25, 28, 29, 37, 38, 40]). The efficacy of substances such as nicardipine [15], aminophylline [20], bisoprolol [24], nifedipine [24], and sodium nitroprusside [28] was evaluated less frequently (one study for each drug). In the case of inhalation anesthetics (desflurane, isoflurane, nitrous oxide), they were mainly applied in control groups [18, 21] or were used only during anesthesia induction and therefore were not subject to the primary analysis of hypotensive efficacy in this review (Table 1).

Notably, significant variability in extreme pressure values was observed. In one study [11], MAP decrease to 44 mmHg was recorded, which exceeded the lower safety limit (50 mmHg) adopted in most protocols. In contrast, several other studies [12, 13, 24] recorded consistently higher-than-intended MAP (exceeding 70–80 mmHg), indicating challenges in achieving hypotensive targets with certain drug protocols (Table 1).

Perfusion monitoring and follow-up period

In all 32 studies analyzed, basic intraoperative monitoring included: blood pressure measurement (including MAP), peripheral oxygen saturation (SpO2), capnography (EtCO2), and ECG recording. In one study [9], this scope was extended to include intraoperative diuresis monitoring and postoperative assessment of laboratory parameters (activated partial thromboplastin time [APTT], hemoglobin [Hgb], hematocrit [Ht], uric acid, electrolytes) (Table 2).

Table 2.

Summary of the intraoperative monitoring methods and patient observation period in the included studies

Studies on a specific drug			Studies comparing specific drugs
Author; year	Intraoperative monitoring	Follow-up observation	Author; year	Intraoperative monitoring	Follow-up observation
Cincikas et al.; 2003 [9]	Basic + diuresis, APTT, Hgb, Ht, urea, electrolytes	Intraoperatively and in the PACU	Skalias et al.; 2023 [40]	Basic	Intraoperatively and in the PACU
Naik et al.; 2011 [10]	Basic	Intraoperatively and in the PACU	Alkan et al.; 2021 [39]	Basic + BIS	Intraoperatively
Ha et al.; 2014 [11]	Basic + TCD via MCA	Intraoperatively	Bharathwaj et al.; 2019 [37]	Basic	Intraoperatively
Heller et al.; 2015 [12]	Basic + NIRS	Intraoperatively and in the PACU	Bajwa et al.; 2016 [36]	Basic	Intraoperatively and in the PACU
Goksu et al.; 2008 [13]	Basic	Intraoperatively and in the PACU(2 h)	Yun et al.; 2015 [22]	Basic + NICOM	Intraoperatively and in the PACU
Sartcaoglu et al.; 2005 [14]	Basic + cognitive tests: MMSE and VADST in 2 h and 24 h	Intraoperatively, in the PACU, and 24 h after surgery	Sajedi et al.; 2016 [23]	Basic	Intraoperatively and in the PACU(2 h)
Zhang et al.; 2021 [15]	Basic + NIRS (rSO2); MMSE; BIS	Intraoperatively and in the PACU; MMSE 24 h after surgery	Sujatha et al.; 2023 [35]	Basic	Intraoperatively and in the PACU
Tirelli et al.; 2004 [16]	Basic	Intraoperatively and in the PACU	Praveen et al.; 2024 [34]	Basic	Intraoperatively, in the PACU, and 24 h after surgery
Li et al.; 2023 [38]	Basic	Intraoperatively, in the PACU, and 7 days after surgery	Bafna et al.; 2021 [33]	Basic	Intraoperatively and in the PACU(0,5 h)
Jiwanmall et al.; 2017 [17]	Basic	Intraoperatively and in the PACU(0,5–1 h)	Sharma et al.; 2025 [32]	Basic	Intraoperatively
Hassan et al.; 2024 [24]	Basic + BIS	Intraoperatively and in the PACU	Escamilla et al.; 2019 [31]	Basic	Intraoperatively, in the PACU, and 7 days after surgery
Gollapudy et al.; 2020 [18]	Basic + FloTrac (CO, CI, SVR, BP)	Intraoperatively and in the PACU	Eghbal et al.; 2018 [30]	Basic	Intraoperatively
Neethirajan et al.; 2020 [19]	Basic	Intraoperatively and in the PACU	Moshiri et al.; 2017 [29]	Basic	Intraoperatively and in the PACU
Rehab et al.; 2023 [20]	Basic	Intraoperatively and in the PACU	Huh et al.; 2020 [28]	Basic + BIS	Intraoperatively and in the PACU
Ankichetty et al.; 2011 [21]	Basic	Intraoperatively, in the PACU, and 48 h after surgery (only PONV and pain)	Karabayirli et al.; 2016 [27]	Basic	Intraoperatively, in the PACU, and 24 h after surgery
			Mugabo et al.; 2024 [26]	Basic	Intraoperatively and in the PACU
			Lee et al.; 2013 [25]	Basic	Intraoperatively and in the PACU

Basic – MAP, HR, ECG, SpO2 and EtCO2

MAP Mean Arterial Pressure, HR Hear Rate, ECG Electrocardiography, SpO2 Peripheral Oxygen Saturation, EtCO2 End-Tidal Carbon Dioxide – Capnography, BIS Bispectral Index, NICOM Non-Invasive Cardiac Output Monitoring, TCD Transcranial Doppler, NIRS Near-Infrared Spectroscopy, rSO2 Regional Cerebral Oxygenation, MMSE Mini–Mental State Examination, VADST Visual Aural Digit Span Test, PACU Post-Anesthesia Care Unit, APTT Activated Partial Thromboplastin Time, Hgb hemoglobin, Ht hematocrit, PONV Postoperative Nausea And Vomiting

Advanced intraoperative monitoring was applied in a total of eight studies [11, 12, 15, 18, 23, 25, 28, 37]. The bispectral index (BIS) (in three studies [15, 25, 37]) and near-infrared spectroscopy (NIRS) (in 2 studies [12, 15]) were used most frequently. In individual cases, the following were used: transcranial Doppler (TCD) assessment of blood flow [11], non-invasive cardiac output monitoring (NICOM) [28], and the FloTrac haemodynamic monitoring system [18]. In two studies [14, 15], the postoperative assessment was supplemented with cognitive function tests: Mini-Mental State Examination (MMSE) and/or Visual Aural Digit Span Test (VADST) (Table 2).

In all reports, patients were continuously monitored intraoperatively, and in most cases, observation also included assessment of clinical status in the post-anesthesia care unit (PACU). Information on further monitoring of patients was not included in 5 studies [11, 25, 26, 32, 35]. The length of stay in the PACU, although rarely specified, ranged from 0.5 to 2 h. Descriptions of postoperative follow-up appeared sporadically and most often covered a period of 24 h [14, 15, 31, 38], 48 h [21] (concerned only pain and postoperative nausea and vomiting (PONV)) or 7 days [22, 34](Table 2). Structured neurological or renal follow-up was infrequent.

Minor and major complications

The safety analysis revealed a varied profile of complications. Among the 32 studies included in the analysis, 14 [9, 10, 14–16, 21, 24, 26, 28, 30, 32, 34, 36, 37] did not report any complications. In the remaining studies, complications were classified as minor or major/hemodynamic (Table 3).

Table 3.

The incidence of minor and major complications reported in the analyzed studies

Studies on a specific drug				Studies comparing specific drugs
Author; year	Used drug	Complications		Author; year	Used drugs	Complications
		Minor	Major			Minor	Major
Cincikas et al.; 2003 [9]	Nitroglycerin	Not reported	Not reported	Skalias et al.; 2023 [40]	Bisoprolol vs. nifedipine	Not reported	Not reported
Naik et al.; 2011 [10]	Propofol TIVA	Not reported	Not reported	Alkan et al.; 2021 [39]	Esmolol vs. remifentanil vs. nitroglycerin	Not reported	10% (REM) and 5% (ESM) of bronchospasm 15% (REM) of bradycardia (requiring atropine)
Ha et al.; 2014 [11]	Propofol TIVA	Not reported	Decrease in Vmca > 50% at MAP < 60 but without complications	Bharathwaj et al.; 2019 [37]	Labetalol vs. dexmedetomidine	Not reported	Not reported
Heller et al.; 2015 [12]	Propofol + remifentanyl + nitrous oxide	11% of PONV in patients with lower SctO2, but not statistically significant	SctO2 <60% in 16% of patients resulted in a longer stay in the PACU	Bajwa et al.; 2016 [36]	Nitroglycerine vs. esmolol vs. dexmedetomidine	6%(DEX), 14% (NIT), and 16% (ESM) of PONV 4% (DEX), 18% (NIT), and 8% (ESM) of shivering 26% (DEX), 8% (NIT), and 6% (ESM) of dry mouth	Not reported
Goksu et al.; 2008 [13]	Dexmedetomidine	13% of PONV	6,7% of bradycardia (requiring treatment with atropine)	Yun et al.; 2015 [22]	Nitroprusside vs. remifentanil	Not reported	Not reported
Sartcaoglu et al.; 2005 [14]	Nitroglycerine	Not reported	Not reported	Sajedi et al.; 2016 [23]	Remifentanil vs. labetalol	9,1% (REM) and 62,1% (LAB) of PONV	3% (REM) and 17,2% (LAB) of hypotension requiring treatment 3% (REM) and 3,4% (LAB) of bradycardia 38% (LAB) of hypotension in PACU
Zhang et al.; 2021 [15]	Nicardipine + esmolol	Not reported	Not reported	Sujatha et al.; 2023 [35]	Dexmedetomidine vs. propofol	Not reported	Not reported
Tirelli et al.; 2004 [16]	Propofol + remifentanyl	Not reported	Not reported	Praveen et al.; 2024 [34]	Dexmedetomidine vs. clonidine	10% (DEX) and 17,5% (CLO) of PONV	15% (DEX) and 10% (CLO) of bradycardia 7,5% (DEX) and 15% (CLO) of hypotension
Li et al.; 2023 [38]	Propofol + remifentanyl	5,5% (3/55) of sore throat; 2% (1/55) of PONV	9,1% (5/55) of hypoxemia;	Bafna et al.; 2021 [33]	Dexmedetomidine vs. clonidine	2,9% (DEX) and 11,4% (CLO) of PONV 11,4% (DEX) and 2,9% (CLO) of dry mouth	8,6% (DEX) and 5,7% (CLO) of bradycardia 8,6% (DEX) and 5,7% (CLO) of hypotension
Jiwanmall et al.; 2017 [17]	Clonidine	6,7% of prolonged sedation	33% of hypotension episodes requiring treatment; 3,3% of bradycardia episodes requiring treatment	Sharma et al.; 2025 [32]	Esmolol vs. dexmedetomidine	Not reported	Not reported
Hassan et al.; 2024 [24]	Dexmedetomidine	5,7% of PONV	3% of bradycardia requiring treatment with atropine	Escamilla et al.; 2019 [31]	Clonidine vs. dexmedetomidine	Not reported	Not reported
Gollapudy et al.; 2020 [18]	Propofol vs. desflurane	8,6% of PONV 10% of prolonged sedation	2,8% of ST changes 4,3% of hypotension	Eghbal et al.; 2018 [30]	Labetalol vs. dexmedetomidine	11% (LAB) of PONV	Not reported
Neethirajan et al.; 2020 [19]	Dexmedetomidine	Not reported	8,7% of bradycardia required treatment with atropine; 13% of hypotension required treatment with ephedrine	Moshiri et al.; 2017 [29]	Propofol vs. dexmedetomidine	Not reported	Not reported
Rehab et al.; 2023 [20]	Aminophylline	11,5% of PONV 7,7% of headache and lightheadedness	7,7% of hypotension 3,9% of bradycardia	Huh et al.; 2020 [28]	Dexmedetomidine vs. remifentanil	Not reported	Not reported
Ankichetty et al.; 2011 [21]	Propofol vs. isoflurane	Not reported	Not reported	Karabayirli et al.; 2016 [27]	Dexmedetomidine vs. remifentanil	21,7% (DEX) and 25% (REM) of PONV 4,2% (both groups) of shivering 13% (DEX) and 21% (REM) of pain required treatment	Not reported
				Mugabo et al.; 2024 [26]	Dexmedetomidine vs. clonidine	Not reported	5% (DEX) of bradycardia required treatment with atropine 5% (DEX) of hypotension required treatment with ephedrine
				Lee et al.; 2013 [25]	Dexmedetomidine vs. remifentanil	Not reported	62% (REM) and 34% (DEX) patients treated with ephedrine 41% (REM) and 25% (DEX) patients treated with atropine

TIVA Total Intravenous Anesthesia, PONV Postoperative Nausea And Vomiting, DEX Dexmedetomidine, REM Remifentanil, CLO Clonidine, LAB Labetalol, ESM Esmolol, NIT Nitroglycerine, SctO2 Cerebral Tissue Oxygenation, MAP Mean Arterial Pressure, Vmca Velocity in the Middle Cerebral Artery

Minor adverse events occurred in 13 studies [12, 13, 17, 18, 20, 22, 23, 27, 29, 31, 32, 35, 38]. The predominant problem was postoperative nausea and vomiting (PONV) reported in 12 studies [12, 13, 18, 20, 22, 23, 27, 29, 31, 32, 35, 38] with a wide range of incidence from 2% to 62%. Less frequently observed complications included prolonged sedation (6.7–10%) [17, 18], postoperative shivering (4–18%) [27, 38], and dry mouth (3–26%) [27, 32]. Individual reports concerned sore throat (5.5%) [22], headaches and dizziness (7.7%) [20], and increased postoperative pain requiring additional analgesia (13–21%) [38] (Table 3).

Significant hemodynamic disturbances and organ complications were reported in 15 studies [11–13, 17–20, 22, 23, 25, 29, 31, 32, 39, 40]. The most frequently reported events were intraoperative bradycardia (< 50–60 beats/min), noted in 10 studies (3–15%) [13, 17, 19, 20, 23, 25, 29, 31, 32, 39], and severe hypotension (defined as MAP < 50 mmHg or SBP < 90 mmHg), reported in 8 studies (4.3–33%) [17–20, 29, 31, 32, 39]. Most of these episodes required pharmacological intervention (e.g., ephedrine or atropine). In one study [29], 38% of patients experienced delayed hypotension in the PACU. Notably, the study by Lee et al. [40] reported high use of rescue medications (ephedrine in 32–64% of patients, atropine in 21–45% of patients), without specifying the corresponding incidence of hypotension or bradycardia (Table 3).

There were also a few clinically critical signs of organ and respiratory ischemia. These included hypoxemia (9.1%) [22], bronchospasm (10%) [25], and ST-segment changes on ECG suggesting myocardial ischemia (10%) [18].

Cerebral perfusion abnormalities included a decrease in cerebral saturation (rScO2) < 60% in 16% of patients (NIRS monitoring) in Heller et al. study (2014) [12], and middle cerebral artery (MCA) flow reduction > 50% with a decrease in MAP < 60 mmHg (TCD) in Ha et al. study [11]. No permanent neurological deficits were reported, although follow-up was insufficient to exclude subtle or delayed complications (Table 3).

Discussion

The available evidence supports the efficacy of controlled hypotension in improving intraoperative conditions during FESS. However, its safety profile remains insufficiently characterized, primarily due to the limited use of advanced monitoring, short and inconsistent postoperative follow-up, and the lack of systematic assessment of organ function (e.g., renal biomarkers, cognitive testing, troponin measurement). Importantly, few studies have directly evaluated the consequences of controlled hypotension or the incidence of complications associated with organ hypoperfusion. Numerous publications have focused on comparing pharmacological agents with respect to surgical field visibility and intraoperative bleeding, whereas safety aspects – particularly the risk of end-organ ischemia – are rarely analyzed in depth. Consequently, the heterogeneity and incompleteness of available data preclude definitive conclusions regarding the safety of controlled hypotension in FESS.

Although the present review was primarily designed to evaluate safety rather than operative efficacy, most of the included studies also reported measures of intraoperative bleeding or surgical field quality. These outcomes are most commonly assessed using the Fromme-Boezaart/Boezaart scale, which grades bleeding intensity and the frequency of suctioning required during endoscopic surgery, and the Wormald surgical field grading scale, a more detailed video-endoscopic instrument developed specifically for assessing bleeding during endoscopic sinus surgery [41, 42]. Operative duration was reported inconsistently across the included studies and, given the safety-focused scope of this review, was not amenable to quantitative synthesis. Overall, the available literature suggests that controlled hypotension may improve surgical field visualization; however, the magnitude and consistency of this benefit vary across studies and should be interpreted alongside the limited evidence regarding end-organ safety.

Furthermore, technique safety is determined not only by the achieved MAP reduction but also by the duration, depth, and frequency of hypotensive episodes. Many studies report average or target MAP values without quantifying hypotension exposure (e.g., time spent below a clinically meaningful threshold, the number of hypotensive events, the rapidity of pressure declines, or the need for rescue vasopressors). This limits the interpretability of the results, since the risk of end-organ ischemia is likely driven by the cumulative hypotension burden rather than by a single intraoperative MAP value. Standardized reporting of hypotension exposure metrics would substantially improve cross-study comparability and enable the identification of non-hazardous boundaries.

Advanced monitoring modalities such as NIRS, TCD, and NICOM remain underutilized, despite providing valuable real-time information on tissue perfusion and oxygenation [43]. Incorporation of such modalities – especially in high-risk individuals – could significantly improve patient safety. Specifically, intraoperative cerebral oximetry enables continuous evaluation of brain oxygenation and facilitates early detection of cerebral hypoperfusion [43]. However, the absence of reported neurological deficits in reviewed studies should not be interpreted as evidence of the absence of cerebral injury. Subclinical cognitive impairments may go undetected without postoperative neuropsychological testing, and the limited duration of follow-up in most studies precludes the identification of delayed neurological sequelae. Future studies should therefore include a structured neurological assessment, ideally consisting of a focused postoperative neurological examination in the PACU and at 24 h, supplemented in research settings by baseline and postoperative cognitive screening. Practical tools could include MMSE or Montreal Cognitive Assessment (MoCA) performed preoperatively and repeated at 24–48 h and, where feasible, at 7–30 days; more sensitive domain-specific tests may be considered when subtle postoperative cognitive dysfunction is a primary endpoint. In addition, in FESS, relatively minor differences in ventilation and EtCO₂ may influence both cerebral perfusion and the surgical field. Hypocapnia can reduce cerebral blood flow through vasoconstriction and might affect cerebral oximetry (NIRS) or TCD-derived flow indices, while CO₂-related changes in vascular tone might also modify bleeding intensity [44]. Therefore, future studies should report – and ideally standardize – ventilatory targets (including a defined EtCO₂ range), as they constitute a plausible confounder for both perfusion monitoring outcomes and surgical field quality.

Regarding cardiac complications, continuous ECG monitoring remains essential; yet it is insufficient to detect all ischemic events. Myocardial Injury after Non-cardiac Surgery (MINS) – a form of prognostically significant myocardial ischemia – may develop without typical ECG changes or clinical symptoms and is often precipitated by intraoperative hypotension-induced oxygen supply-demand mismatch [45, 46]. Consequently, routine postoperative assessment of cardiac injury biomarkers, particularly high-sensitivity cardiac troponin T (hs-cTnT), should be considered in patients at increased cardiovascular risk or in those exposed to prolonged or profound hypotension [47]. This risk-stratified approach is consistent with contemporary perioperative guidelines, including the 2017 Canadian Cardiovascular Society recommendations and the 2022 European Society of Cardiology guidelines, which advocate troponin surveillance in patients at increased cardiovascular risk undergoing major non-cardiac surgical procedures [48, 49]. Importantly, MAP is only a surrogate for organ perfusion and does not directly reflect forward flow or oxygen delivery. Perfusion depends on cardiac output/cardiac index and systemic vascular resistance, so two hypotensive protocols achieving the same MAP may generate markedly different oxygen delivery (DO₂) and, consequently, distinct risks of MINS or acute kidney injury (AKI) [50]. This supports the selective use of hemodynamic monitoring, including flow parameters (e.g., NICOM, FloTrac), particularly in patients with limited cardiovascular reserve.

Although none of the reviewed studies systematically assessed renal function, MAP below 65 mmHg is widely recognized as a significant predictor of AKI. The absence of postoperative creatinine follow-up in the analyzed studies represents a considerable knowledge gap, especially for vulnerable patients with pre-existing renal disease, dehydration, or treatment with angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs). These populations are particularly prone to hypoperfusion-related injury. From a pragmatic and cost-conscious perspective, routine advanced biomarker surveillance is unlikely to be justified in all low-risk FESS patients. However, serum creatinine assessment within 24–48 h is relatively inexpensive and should be considered in high-risk patients, after prolonged or deep hypotension, or when intraoperative rescue vasopressors are required. Future research should integrate perioperative measurements of renal biomarkers, such as Neutrophil Gelatinase-Associated Lipocalin (NGAL) and cystatin C, together with urine output assessment, to better define the renal safety margins of controlled hypotension [51].

Another critical limitation of the reviewed literature concerns the wide variability in pharmacological regimens applied to induce hypotension. The lack of standardized protocols across studies hinders direct comparison of safety profiles. Furthermore, essential perioperative variables – including preoperative hydration status, intraoperative fluid balance, and the continuation or withdrawal of ACEI-ARB therapy – were rarely reported. This omission is crucial, as both preoperative hypovolemia and the blunted renin-angiotensin response in ACEI/ARB-treated patients may potentiate drug-induced hypotension, leading to profound and resistant decrease in blood pressure. Ensuring adequate hydration and establishing standardized guidelines for perioperative management of antihypertensive therapy are therefore imperative steps toward reducing the risk of end-organ hypoperfusion.

Heart rate control constitutes an additional clinically relevant determinant of surgical field quality and systemic organ perfusion. Vasodilator-based hypotensive techniques may elicit baroreceptor-mediated reflex tachycardia and increased myocardial contractility, thereby maintaining or increasing cardiac output, enhancing peripheral vasodilatation, and potentially contributing to venous oozing within the operative field. Consequently, in routine anesthetic practice for FESS, optimal surgical conditions are often achieved not through MAP reduction alone, but through a combined strategy of moderate controlled hypotension and concomitant heart rate control, commonly with a target heart rate of approximately 60–70 beats/min, using beta-blockers such as esmolol or sympatholytic agents such as dexmedetomidine [25, 27, 42]. This interaction is clinically significant, as both arterial hypotension and pharmacological heart rate reduction may independently, and in combination, decrease oxygen delivery, particularly in patients with limited cardiovascular reserve. Future studies should therefore systematically report heart rate targets, use of beta-blockers and other negative chronotropic agents or sympatholytic agents, requirements for vasopressor rescue, and cardiac output or other flow-related parameters, rather than relying on MAP values alone.

Most authors have adopted a target MAP of 50–70 mmHg as the operational range for controlled hypotension during FESS. Although MAP values below 50 mmHg may approach or exceed the lower limits of cerebral and renal autoregulation, particularly in patients with hypertension or established vascular disease, MAP values above 70–80 mmHg may compromise surgical field visibility, prolong operative time, and potentially increase procedure-related risk [50]. Accordingly, controlled hypotension should be regarded as an individualized risk-benefit intervention, with hemodynamic targets adjusted accordingly to each patient’s baseline blood pressure, cardiovascular reserve, and comorbidity profile. We do not consider it appropriate to recommend a single universal percentage reduction from baseline for all patients undergoing FESS. A pragmatic approach would be to report both absolute MAP values and relative reductions from preoperative baseline, to avoid MAP values below 50 mmHg in all patients, to avoid sustained MAP values below 65 mmHg in higher-risk individuals, and, in general, to limit MAP reduction to no more than approximately 20–30% from baseline. In patients with chronic hypertension, diabetes mellitus, atherosclerosis, renal impairment, or other forms of vascular disease, even more conservative reductions may be warranted, as autoregulatory thresholds may be shifted toward higher perfusion pressures [52–54]. Reporting and interpreting relative hypotension in conjunction with absolute MAP thresholds may therefore provide a more physiologically appropriate framework and may facilitate the development of patient-specific, dynamically guided hypotensive protocols, as has been proposed in other surgical disciplines.

Furthermore, current research on optimal MAP targets in FESS is limited by small sample sizes, heterogeneous study designs, and inconsistent reporting of complications. The diversity of pharmacological agents and the lack of standardized definitions for adverse events impede robust cross-study comparisons. Few studies incorporate advanced perfusion monitoring, which limits the detection of organ-specific disturbances and restricts interpretation to MAP rather than actual tissue oxygen delivery. Moreover, many trials omit key parameters such as hypotensive duration, frequency of perfusion disturbances, or postoperative biomarker trends, rendering their conclusions potentially misleading.

Future research should aim to define both standardized and patient-individualized MAP targets for controlled hypotension in FESS. These studies should integrate multimodal monitoring beyond basic hemodynamic measures, combining MAP assessments with markers of cerebral, renal, and myocardial perfusion. In addition, protocols should quantify hypotension exposure (e.g., time below thresholds, number and depth of hypotensive episodes) and relate achieved pressures to baseline hemodynamics to clarify the role of relative hypotension in high-risk subgroups. Routine postoperative evaluation of serum troponin and creatinine levels, as well as cognitive function, would allow for a more comprehensive assessment of the physiological consequences and long-term safety of controlled hypotension. Only such an approach will enable the establishment of validated safety thresholds and the development of personalized hemodynamic management strategies in endoscopic sinus surgery (Fig. 7).

Fig. 7.

The primary issues and proposed solutions. MAP – Mean Arterial Pressure

Conclusions

Controlled hypotension markedly improves intraoperative conditions during FESS; however, its systemic safety, particularly with respect to organ perfusion, remains incompletely defined. Current evidence primarily addresses surgical visibility and bleeding reduction, while subclinical complications (cerebral desaturation, myocardial injury, AKI) may be overlooked without appropriate intraoperative monitoring and postoperative biomarker surveillance. The widespread application of a uniform MAP target of 50–70 mmHg may be insufficient, particularly in older individuals with cardiovascular or metabolic comorbidities. A paradigm shift toward individualized hemodynamic management, tailored to baseline blood pressure, heart rate, comorbidities, and overall risk profile, is required. Selective, risk-stratified use of perfusion-oriented monitoring (NIRS, NICOM, TCD) and routine postoperative renal/cardiac biomarker assessment, together with standardized complication reporting and longer follow-up (including neurological outcomes), is needed to define safe limits for controlled hypotension in FESS.

Authors' contributions

Conceptualization: M.A, S.B. Methodology: M.A. Data Curation: R.K., L.Z., W.SG. Formal Analysis: M.A., J.D., R.K., L.Z., W.SG. Writing – Original Draft: M.A., R.K. Writing – Review & Editing: M.A., J.D., R.K., L.Z., W.SG. Visualization: L.Z., W.SG. Supervision: M.A., S.B. All authors have approved the final version of the manuscript.

Funding

No funding was received to assist with the preparation of this manuscript.

Data availability

The data generated during this study are available within the article. Datasets analyzed during the current study preparation are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.