The effect of oral stimulants on thirst of patients in the post-operative period: a systematic review and meta-analysis of randomized controlled trials Authors: Sogand Sarmadi, Fateme Monjazebi, Neda Sanaie, Akbar Zare-Kaseb, Haniyeh Ghanbari Nasab, Sara Beigzadeh, Amir Mohamad Nazari Published: 2025 ABSTRACT AbstractIntroductionThirst is a prevalent symptom during the immediate postoperative period. It appears that the utilization of oral stimulants is a primary method for managing these symptoms. This study aimed to conduct a systematic review of studies that assessed the impact of oral stimulants on patients’ thirst during the immediate post-operative period.MethodA thorough search was performed on six electronic databases, including PubMed, Web of Science, Scopus, Cochrane Library, CINAHL, and Embase, from each database’s earliest inclusive dates to 17 September 2025. No limitations were placed on the language of the studies included in this study. The risk of bias assessments were executed utilizing the Cochrane Risk of Bias 2 tool. A meta-analysis was executed, with additional subgroup analysis. The publication bias was examined using funnel plots and Begg and Egger’s test. A sensitivity analysis was conducted utilizing the leave-one-out approach. The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach was employed to assess the certainty of the evidence.ResultsData were collected from 14 randomized clinical trials (RCTs) with 1289 patients for meta-analysis. Our meta-analysis showed that the pooled SMD for thirst intensity difference across groups was − 1.38 (95% CI: -1.92, -0.84). Included studies exhibited significant heterogeneity (I2 = 94.81%). The subgroup analyses in two subgroups (starting point for evaluations and baseline inequality) justified the high heterogeneity. The effect was derived from three oral stimulants (menthol, cold water, and ice), and when analyzed individually, menthol and cold water reached significance in reducing thirst intensity compared with controls. Moreover, our study has highlighted the likelihood of publication bias for thirst intensity. The results of the GRADE evaluation demonstrated a low level of evidence regarding the severity of thirst.ConclusionBased on our study’s findings, interventions involving menthol, cold water, and ice have significantly decreased post-surgical thirst intensity. Obtaining more robust conclusions requires that subsequent research adopt standardized methodological practices, thereby reducing methodological heterogeneity.RegistrationThe study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) at the National Institute for Health Research (No: CRD42024599788).Supplementary InformationThe online version contains supplementary material available at 10.1186/s12903-025-07249-8.Keywords: Oral stimulants, Thirst, Post-operative period, Menthol Introduction Postoperative thirst, a prevalent psychological symptom, can be influenced by various environmental and individual factors, including anesthetic drugs, opiates (which reduce saliva secretion), and blood loss resulting from invasive procedures during surgery, fasting before surgery, intubation during and after surgery, and dehydration [1, 2]. The continuous shortage of water can have detrimental impacts on the cognitive and psychological abilities of patients [3, 4]. Following the surgical procedure, patients indicated that thirst held greater significance than hunger; in certain instances, this sensation exceeded physical pain in intensity. Research findings have shown that the prevalence of thirst in postoperative patients varies from 18.5% to 97.6% in certain studies [5]. According to various studies, the frequency of this symptom during the immediate postoperative period can be as high as 78% [1]. The sensation of thirst is characterized as the urge to consume water or other fluids, serving as a stimulus for the body’s necessary liquid intake [6]. It is crucial to distinguish between the sensations of thirst, xerostomia, and thirst discomfort. Xerostomia, often known as dry mouth, is the medical term for the subjective feeling of oral dryness, frequently resulting from decreased salivary flow (hyposalivation) [7]. Thirst discomfort entails the affective, cognitive, and behavioral dimensions of the thirst experience, such as distress, urgency, preoccupation, and the motivational impulse to obtain fluids [8]. Although they frequently appear together, the evidence supports the notion that these symptoms are separate but interconnected outcomes that should be evaluated independently in patients after surgery [9, 10]. Hypovolemic thirst is a consequence of hypothalamic-pituitary endocrine reactions [11]. In response to hyperosmolarity, the perception of thirst is attributed to the osmotic receptors present in the hypothalamus. Consequently, the hypothalamus releases antidiuretic hormone through signaling mechanisms [12]. In the event of hypovolemia during surgery, such as excessive bleeding, there is a significant reduction in plasma volume, resulting in a subsequent decline in arterial blood pressure [13]. Activating the renin-angiotensin-aldosterone system (RAAS) results in a decrease in renal blood flow and the stimulation of renin secretion. Renin facilitates the conversion of angiotensinogen into angiotensin I, which is subsequently transformed into angiotensin II through enzymatic catalysis [14, 15]. Angiotensin II functions as a potent dipsogenic hormone by directly stimulating angiotensin II-sensitive neurons in the subfornical organ (SFO) and other circumventricular structures, thus heightening thirst perception and promoting fluid-seeking behavior [16]. Thirst perception is elicited by transmitting water-seeking signals to the SFO and the terminal vasculature organ in the central nervous system [17]. The processes of satiety before absorption and satiety after absorption are two mechanisms that aid in quenching thirst after the intake of water [18]. Cold water or ice can stimulate Transient Receptor Potential Melastatin 8 (TRPM8), cold receptors in the oropharynx, resulting in pre-absorptive satiety [19]. Post-absorptive satiety is triggered by water absorption from the digestive tract and the detection of reduced blood osmolarity by osmoreceptors in the hypothalamus. The state of pre-absorptive satiety is experienced before post-absorptive satiety [18]. Prior research has examined diverse approaches, such as using cooling agents, ice, menthol, chewing gum, acupressure, and the conventional technique of employing wet cotton swabs to alleviate thirst [20, 21]. Menthol is an emerging agent for relieving postoperative thirst by activating afferent pathways impacting the sensory, orbitofrontal, and cingulate cortices [22]. Ice is characterized as the solid form of water, formed when the temperature falls below 0 degrees Celsius, and cold water, typically ranging from 5 to 10 degrees Celsius above its freezing point [5]. According to the findings of the studies, the consumption of cold liquids appears to be more advantageous than the consumption of liquids at room temperature in terms of reducing thirst, facilitating the transmission of inhibitory signals from SFO thirst neurons, and increasing saliva secretion [23, 24]. Given the importance of postoperative thirst in patients, the insufficient understanding among clinicians regarding its management, and the scarcity of comprehensive systematic reviews and meta-analyses on the effectiveness of oral stimulants, we have opted to conduct a systematic review and meta-analysis to address this issue. This study distinguishes itself from other systematic reviews by focusing on oral stimulants such as menthol, cold water, and ice and evaluating interventional studies with random allocation. Also, our study used Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) scores to assess the certainty of results regarding each outcome. Thus, we aimed to evaluate oral stimulants’ effectiveness on patients’ thirst in the postoperative period. Method Our systematic review protocol has been reported following the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement [25]. To uphold transparency and accountability, the protocol for this review was formally registered with PROSPERO under the identification number CRD42024599788. Formulating the review question The review question of this systematic review is formulated according to PICO(S) components [26]. The review question is as follows: “Are postoperative oral stimulants effective in relieving the thirst of adult patients after surgery?“. Population Any studies on immediate post-operative evaluations of surgical patients admitted to post-anesthesia, intensive care units (ICU), and surgery wards. Interventions Oral stimulants (Cold water, Ice, and Menthol) were among the agents used as interventions in primary studies to be included. Comparators Standard care (Water at room temperature or a non-menthol package) was the only eligible comparator for inclusion. Outcomes Thirst (primary outcome) intensity measured by the Numeric Rating Scale (NRS) and Visual Analogue Scale (VAS) was the desired outcome. Thirst discomfort and dry mouth measured by self-reported questionnaires, thirst measured by other objective tests, and incidence and prevalence of thirst were all eligible for study inclusions. Study design This study only included randomized controlled trials (RCTs). Literature search We searched six electronic databases: PubMed, Web of Science, Scopus, Cochrane Library, CINAHL, and Embase. In this study, we did not impose any restrictions on the language of the included studies. We also searched the gray literature, including conference abstracts and theses. We checked the reference list of included articles and relevant review studies to ensure all eligible articles were included. We performed a comprehensive search of all databases, starting from the initial accessible date and continuing until September 27, 2024. Considering the time needed to complete the article, the search was revised to include the timeframe up to September 17, 2025. Searches were performed using two approaches: intervention and population or intervention and outcomes. The strategy with the desired Number Needed to Read (NNR) was selected for each database. To reduce the possibility of missing studies, searches were conducted as widely as possible. Mesh, Emtree, free-text method, and the expert’s opinion were used for each search component. Proper Boolean operators and database filters were applied to optimize the search. The strategies were peer-reviewed by another author (S.S.) before execution using the Peer Review of Electronic Search Strategies Checklist [27]. Google Scholar and ClinicalTrials.gov were also searched in September 2025 to identify any registered yet unpublished or ongoing clinical trials. Finally, we contacted subject-area experts for information about ongoing studies. The Supplementary Appendix presents the detailed search strategy. Selection process EndNote, a reference management software, was used to import the titles and abstracts of the relevant studies, and then any duplicates were eliminated. Subsequently, records were assessed by title and abstract, filtering out irrelevant citations according to the inclusion and exclusion criteria. The full texts of the possibly relevant studies were obtained and evaluated by two researchers (A.Z. and S.S.) separately. A third researcher (A.N.) discussed and resolved study selection conflicts. In instances where there are multiple publications, such as posters, preprint publications, and theses, we have only considered the published paper. Data extraction Pairs of authors (S.S. and A.Z., H.GH., and S.B.) independently extracted data on the characteristics of the eligible studies. A Microsoft Excel spreadsheet (Microsoft, Redmond, Washington, USA) was employed and devised before the extractions (refer to online supplemental appendix 2). The extracted data was reviewed by a third author (A.N.) to identify any errors or disagreements. Risk of bias assessment Each included study was independently assessed for risk of bias by two authors (A.Z. and S.S.). In the event of any disagreements, the two authors engaged in discussion and sought the input of a third author (N.S.) for final evaluation in cases where consensus could not be reached. The risk of bias assessment in the primary studies was performed using V.2 of the Cochrane risk of bias tool for randomized trials (Cochrane RoB 2 tool). The Cochrane RoB 2 tool evaluates five domains of bias: [1] selection bias [2], performance bias [3], attrition bias [4], detection bias, and [5] selective outcome reporting bias. Data synthesis and analysis Thirst outcomes were standardized and assessed as a standardized mean difference (SMD) and mean difference (MD) across all studies. The calculation of SMD was performed using Hedges` g, according to the conditions of the primary studies. SMD interpretation was as follows: <0.2 = no/trivial effect; 0.2 to 0.5 = small effect; >0.5 to 0.8 = medium effect; >0.8 = large effect [28]. The means and standard deviations were calculated using the method outlined by Wan et al. [29], when medians and interquartile ranges were given. The current meta-analysis revealed that outcome measures in some studies had a mean/SD ratio of less than 2, indicating the presence of skewness [30]. However, the lack of normality did not pose a significant problem, as the majority of studies had an adequate sample size (i.e., ≥ 30 in each group), and thus, the normality of SMD estimates was justified by invoking the central limit theorem [22]. Some studies reported endpoint data only; thus, we couldn’t calculate the mean change for some of the included studies. We used endpoint data and implemented a subgroup analysis regarding baseline inequality to address the challenge of using this meta-analysis approach. The utilization of forest plots allowed for the depiction of effect sizes and their respective confidence intervals (CI) for each study grouping. The random-effects model was employed for all meta-analyses utilizing the restricted maximum likelihood (REML) model. This decision was made due to the anticipated significant heterogeneity between studies, such as variations in intervention parameters and co-interventions, as well as the limited number of studies in each group. The number needed to treat was also calculated using the Psychometrica online calculator (http://www.psychometrica.de/effect_size.html) as an estimate of the clinical benefit for the primary outcomes. Meta-analyses were conducted using Stata Multiprocessor, Version 17.0 (StataCorp, College Station, Texas, USA). Assessment of heterogeneity The I2 statistic (inconsistency index) and Q test (χ2) were employed to quantify the statistical heterogeneity among the included studies, with a significance level set at p ≤ 0.050 [31]. The interpretation of I2 values was as follows: 0%–40%: heterogeneity may not be important; 30%–60%: may represent moderate heterogeneity; 50%–90%: may represent substantial heterogeneity and 75%–100%: considerable heterogeneity [32]. Subgroup analysis To investigate potential sources of heterogeneity, we evaluated the differences among studies based on the following factors: (i) population characteristics, including geographical location and age groups, (ii) risk of bias of the included studies, divided by low, moderate, and high risks of bias in total and each domain (iii) fasting duration (< 12 h, 12–15 h, > 15 h), intervention type (Cold water, Ice, Menthol), surgical procedure, baseline inequalities, and starting point of outcome evaluations (from end of surgery, extubation, of first intervention). Assessment of publication bias We utilized various techniques to assess publication bias, such as funnel plot analysis, Begg’s and Egger’s statistical tests, and the Trim and Fill method. Moreover, for Begg’s and Egger’s tests, publication bias was determined if at least one of the tests achieved a significant level. The test’s significance level was considered to be 0.05. If the 95% confidence interval for the slope of the linear regression encompasses zero, it suggests the absence of publication bias. Sensitivity analysis This systematic review and meta-analysis utilized three approaches for sensitivity analysis. First, we performed a sensitivity analysis using the leave-one-out-remove method to assess the consequences of excluding individual studies from the overall estimates. Second, a sensitivity analysis was conducted by excluding high-risk studies regarding risk of bias assessments. A third sensitivity analysis approach was performed by considering third groups from three-group studies whose third-line interventions were oral cold stimuli to identify changes in the results. Certainty of evidence We examined the certainty of evidence for the primary outcomes using the GRADE approach [33]. The implemented criteria are fully presented in the supplementary appendix. Result Study selection A total of 3,742 records were identified through electronic database searches (PubMed = 193; Web of Science = 666; Scopus = 1,406; Cochrane Library = 189; CINAHL = 313; Embase = 975), with no identified records from registries. After removing 2,314 duplicate records, 1,428 remained for title and abstract screening. During this process, 1,398 records were excluded (512 were not aligned with the research question, and 886 were case reports, editorial letters, descriptive studies, dissertations, reviews, or other ineligible study designs), leaving 30 records for which full texts were sought. Four full texts could not be retrieved, so 26 full-text articles were assessed for eligibility. Of these, 12 were excluded after full-text review: 7 for inappropriate study design or outcomes and 5 for lacking the desired information regarding the primary outcome in a form suitable for meta-analysis. In addition, one record was identified through citation searching, but it was assessed and excluded for inappropriate design. Ultimately, 14 studies met the inclusion criteria and were included in the quantitative synthesis (meta-analysis; n = 14) (Fig. 1) [2, 6, 24, 34–44]. PRISMA Flowchart Characteristics The sample size of the included studies varied from 44 to 208. A total of 1289 participants were included in this review. All of the included studies were RCTs. Two distinct studies were undertaken, focusing specifically on patients undergoing cardiac surgery [37, 42]. Four studies examined different approaches to abdominal surgery [6, 34, 36, 44], while one focused on radical prostatectomy [35]. Seven studies were conducted on patients who underwent various surgical procedures [2, 24, 38–41, 43]. Five studies used cold water as the primary intervention [24, 34, 36, 37, 43]. Menthol was used in six studies [6, 35, 38, 39, 42, 44]. Also, three studies used ice [2, 40, 41]. Thirst severity was reported in all studies, while only three reported thirst discomfort [35, 37, 38]. The characteristics of the included studies are summarized in Table 1. Table 1Basic characteristics of the included studies in this systematic reviewFirst Author/yearStudy characteristics1. Design2. Sample Size (I/C)ParticipantsMean Age (SD)Measured Outcomes (measurement tools)InterventionComparisonAslan et al. 2025 [44]1. RCT2. 159 (54/53/53)General surgery and liver transplant intensive care units patientsIG1: 36.41 ± 14.44IG2: 41.79 ± 16.35CG: 39.05 ± 13.731.Numerical rating scale (NRS) (0–10)IG1: Menthol gumIG2: Frozen salineThe control group did not receive any intervention.Eren et al. 2025 [2]1. RCT2. 150 (50/50/50)Patients transferred to the PACU after surgeryIG1: 41.86 ± 14.24IG2: 38.24 ± 12.38CG: 42.58 ± 14.361.Visual Analogue Scale (VAS) (0–10)IG1: WaterIG2: IceThe control group did not receive any intervention. Rajeshkumar et al. 2025 [43]1. RCT2. 60 (30/30)General surgery patientsNA1. Numerical Rating Scale for Thirst (NRS)Sterile cold swabs two and cold water sprayRoutine care Gungor et al. 2024 [34]1. RCT2. 110 (55/55)Abdominal surgery intensive careunit patientsIG: 53.20 ± 10.29CG: 50.90 ± 14.521.Numerical rating scale (NRS)Cold water sprayed into the patient’s mouth three times per hour.The control group did not receive any intervention. Lian et al. 2024 [24]1. RCT2. 56 (28/28)Surgical intensive care unit patientsIG: 52.82 ± 17.43CG: 55.89 ± 14.491.Numeric rating scale (NRS)Application of ice-coldwater to manage thirstMoistening the oral cavity with a cotton swab dipped in sterilized injection water. Mert et al. 2024 [6]1. RCT2. 90 (30/30/30)Laparoscopic cholecystectomy patientsIG1: 47.13 ± 12.25IG2: 48.23 ± 10.42CG: 47.43 ± 13.581.Visual Comparison Scale (VCS)IG1: Menthol iceIG2: Ice popsiclesPatients in the control group did not undergo any intervention. Can et al. 2023 [42]1. RCT2. 119 (59/60)Cardiac surgeryIGz,2: 58.0 ± 5.3z,2 Intervention groupy,3 Control groupCGy,3: 56.9 ± 7.51.Visual Analogue ScaleMenthol lozengeRoutine clinical procedures da Silva et al. 2023 [35]1. RCT2. 44 (22/22)Radical prostatectomyIG: 66.4 ± 7.4CG: 67.0 ± 6.61.Thirst Intensity: Numerical Scale (NS)2.Thirst discomfort: Perioperative Thirst Discomfort ScaleThe filtered water popsicle with the addition of 0.05% menthol substrate.A filtered water popsicle without the addition of 0.05% menthol substrate. Lin et al. 2022 [37]1. RCT2. 165 (96/49)Cardiac surgery patientsCG: 56.57 (13.32)IG A: 54.43 (16.21)IG B: 55.98 (15.02)1.Thirst Intensity: Numerical Scale (NS)2.Thirst discomfort: Perioperative Thirst Discomfort ScaleSpray-based oropharyngeal moisturizing programUsual careOztas et al. 2022 [36]1. RCT2. 104 (34/35/35)Major Abdominal SurgeryIG1: 61.44 ± 10.39IG2: 63.48 ± 13.43CG: 58.02 ± 12.631.Visual Analog Scale (VAS)IG1: Cold water spray orallyIG2: Cold saline (0.9% NaCl) spray orallyWater at room temperature using a syringe. Conchon et al. 2021 [38]1. RCT2. 50 (25/25)Elective SurgeryMedian =IG: 66 (63–71)CG: 69 (64–77)1.Thirst Intensity: Numerical Scale (NS)Thirst discomfort: Perioperative Thirst Discomfort Scale Discomfort Scale)A 20 ml mentholated popsicle.Usual care Serato et al. 2019 [39]1. RCT2. 120 (59/61)Anesthetic Recovery Room PatientsIG: 37.4 (11.0)CG: 39.5 (11.8)1.Numerical Verbal Scale (NVS)Package of menthol measures (made of a 10 ml popsicle, ultrafiltered water, 0.05% menthol, and lip moisturizing with 1% menthol).Package of non-menthol measures (ultrafiltered water-based and hydrating popsicle with the same formulation as the experimental group without the addition of menthol). Conchon et al. 2018 [40]1. RCT2. 208 (104/104)Patients Undergoing Elective or Urgent SurgeryEG: 39.3 ± 13.9CG: 36.4 ± 12.61.Numeric rating scale (NRS)10-mL ice popsicle.10 mL of water at room temperature. Aroni et al. 2012 [41]1. RCT2. 90 (45/45)Post-Anesthetic PatientsNA1.Numeric rating scale (NRS)Administration of iceAdministration of water Basic characteristics of the included studies in this systematic review 1. RCT 2. 159 (54/53/53) IG1: 36.41 ± 14.44 IG2: 41.79 ± 16.35 CG: 39.05 ± 13.73 IG1: Menthol gum IG2: Frozen saline 1. RCT 2. 150 (50/50/50) IG1: 41.86 ± 14.24 IG2: 38.24 ± 12.38 CG: 42.58 ± 14.36 IG1: Water IG2: Ice 1. RCT 2. 60 (30/30) 1. RCT 2. 110 (55/55) Abdominal surgery intensive care unit patients IG: 53.20 ± 10.29 CG: 50.90 ± 14.52 1. RCT 2. 56 (28/28) IG: 52.82 ± 17.43 CG: 55.89 ± 14.49 Application of ice-cold water to manage thirst 1. RCT 2. 90 (30/30/30) IG1: 47.13 ± 12.25 IG2: 48.23 ± 10.42 CG: 47.43 ± 13.58 IG1: Menthol ice IG2: Ice popsicles 1. RCT 2. 119 (59/60) IGz,2: 58.0 ± 5.3 z,2 Intervention group y,3 Control group CGy,3: 56.9 ± 7.5 1. RCT 2. 44 (22/22) IG: 66.4 ± 7.4 CG: 67.0 ± 6.6 1.Thirst Intensity: Numerical Scale (NS) 2.Thirst discomfort: Perioperative Thirst Discomfort Scale 1. RCT 2. 165 (96/49) CG: 56.57 (13.32) IG A: 54.43 (16.21) IG B: 55.98 (15.02) 1.Thirst Intensity: Numerical Scale (NS) 2.Thirst discomfort: Perioperative Thirst Discomfort Scale 1. RCT 2. 104 (34/35/35) IG1: 61.44 ± 10.39 IG2: 63.48 ± 13.43 CG: 58.02 ± 12.63 IG1: Cold water spray orally IG2: Cold saline (0.9% NaCl) spray orally 1. RCT 2. 50 (25/25) Median = IG: 66 (63–71) CG: 69 (64–77) 1.Thirst Intensity: Numerical Scale (NS) Thirst discomfort: Perioperative Thirst Discomfort Scale Discomfort Scale) 1. RCT 2. 120 (59/61) IG: 37.4 (11.0) CG: 39.5 (11.8) 1. RCT 2. 208 (104/104) EG: 39.3 ± 13.9 CG: 36.4 ± 12.6 1. RCT 2. 90 (45/45) Risk of bias assessment In terms of sequence generation, all studies that used random sequences (e.g., generated by a computer program, coin toss, etc.) were classified as ‘low-risk,’ except for the study by Rajeshkumar et al. 2025 [43], which did not describe the randomization process in detail, potentially raising concerns. Seven studies were deemed ‘low-risk’ regarding performance bias for the characteristics of the intended program [6, 34–38, 44]. Others were evaluated as having ‘Some concerns’ or ‘High-risk.’ Regarding incomplete data, three [37, 42, 43] studies using per-protocol analyses were assessed as ‘Some concerns.’ All studies except five were regarded as ‘low-risk’ in the reporting bias domain because they reported all outcome measures [6, 36, 37, 39, 43]. The results of the risk of bias assessment for the included studies are shown in Figs. 2 and 3. Risk of bias graph of the included RCT studies Risk of bias summary of the included RCT studies Cold oral stimulants and thirst intensity Cold oral stimulants significantly reduced thirst intensity (P < 0.001). The pooled SMD and 95% CI were − 1.38 (−1.92, −0.84), which favored the intervention groups. Regarding heterogeneity among the estimated SMD for each included study, there was considerable heterogeneity among studies (I2 = 94.81%) (see Fig. 4). The 95% Prediction Interval (PI) was estimated at −3.64 to 0.87, which shows the probable estimates for future studies in this field. Also, it was found that cold oral stimulants yielded a 2.30 units reduction in thirst intensity on the 0–10 NRS or VAS scale (P < 0.001). Forest plot for pooled Standardized Mean Difference of included studies The effectiveness of these interventions was significantly different among different types of surgeries (P = 0.01) and agents used (P < 0.001). Thus, cardiac surgery patients were more affected considerably by desired interventions than other surgeries (SMD: −2.53, (95% CI: −3.16, −1.89)). Also, cold water was more effective in relieving the thirst of patients (SMD: −1.78, (95% CI: −2.49, −1.07)). Studies conducted in Asia showed higher effectiveness when compared to South America (SMD=−1.70 vs. −0.80, P = 0.09). Regarding the risk of bias, all domains except for the selection of the reported data contributed to a significant difference in estimates of SMD. For all domains except performance bias, studies with lower quality revealed overestimated effects compared to low-risk studies. Intervention efficacy was significantly higher in studies where patients underwent 12–15 h of fasting before the procedure, compared with other fasting durations. Also, the effectiveness of interventions in studies with a mean age of less than fifty years has been significantly higher than those with a mean age of more than fifty years (p < 0.001). Considering the two primary limitations of this study, namely the variations in starting point for evaluations (P = 0.26) and the disparities in baselines (P = 0.42) across several studies, it is worth noting that neither of these factors had a meaningful effect on the calculated effect size (Table 2). Table 2Sub-group results of factors contributing to variations in estimated SMD among included studiesVariablesVariables CategoriesNo. of included studies in each sub-categorySub-group Pooled SMD [95% CI]Heterogeneity Across studies [I2]Test of group differencesType of SurgeryAbdominal Surgery4-1.41(-1.99, -0.83)83.180.01Cardiac Surgery2-2.53(-3.16, -1.89)68.62Radical Prostatectomy1-1.23(-1.87, -0.60)-Not Specified7-1.06(-1.96, -0.16)96.66 Type of AgentCold Water5-1.78(-2.49, -1.07)88.86<0.001Ice3-0.24(-0.72, 0.24)81.90Menthol6-1.65(-2.49, -0.81)93.85ContinentAsia9-1.70(-2.34, -1.06)93.300.09South America5-0.80(-1.62, 0.02)94.59CountryBrazil5-0.80(-1.62, 0.02)94.59<0.001China1-1.60(-2.19, 1.00)-India1-3.06(-3.80, -2.31)-Turkey7-1.54(-2.26, -0.82)94.05Starting Point for EvaluationsFrom Extubation2-2.24(-3.46, -1.01)89.850.26From Surgery2-1.13(-1.50, -0.76)26.23From the 1st Intervention8-1.07(-1.81, -0.34)95.88Not Specified2-2.13(-3.92, -0.35)92.55Tool TypeNRS (0-10)10-1.32(-1.93, -0.72)94.360.76VAS (0-10)4-1.54(-2.82, -0.26)96.23D1 ROBLow13-1.26(-1.78, -0.74)94.29<0.001Some Concerns1-3.06(-3.80, -2.31)-D2 ROBHigh1-0.69(-0.97, -0.41)-<0.001Low7-1.52(-1.92, -1.11)74.32Some Concerns6-1.34(-2.54, -0.14)97.60D3 ROBLow11-1.02(-1.51, -0.53)92.59<0.001Some Concerns3-2.66(-3.18, -2.15)57.96 D4 ROBHigh1-3.06(-3.80, -2.31)-<0.001Low9-1.28(-1.94, -0.63)95.08Some Concerns4-1.21(-2.19, -0.24) 93.26D5 ROBLow12-1.31(-1.87, -0.75)93.950.66Some Concerns2-1.85(-4.16, 0.47)97.08Overall, ROBHigh2-1.85(-4.16, 0.47)97.080.84Low4-1.44(-1.88, -1.01)64.88Some Concerns8-1.23(-2.05, -0.42)95.89Fasting Time for Liquid (Hours)<12 h1-0.04(-0.43, 0.35)-<0.00112-15 h2-1.81(-2.88, -0.74)86.96>15h5-1.05(-1.75, -0.35)92.52Not Specified6-1.74(-2.70, -0.78)95.07Fasting Time for Solid Foods (Hours)<12 h1-0.04(-0.43, 0.35)-<00112-15 h1-2.39(-3.05, -1.73)->15 h6-1.09(-1.66, -0.51)90.94Not Specified6-1.74(-2.70, -0.78)95.07Mean Age<505-0.85( -1.67, -0.02)95.55<0.001>507-1.76(-2.28, -1.24)84.99Not Specified10.08(-0.33, 0.49)-Baseline InequalityYes3-0.93(-2.80, 0.94)98.220.42No9-1.60(-2.22, -0.98)93.17Not Specified2-1.13(-1.50, -0.76)26.231st, first; ROB, Risk of Bias; D, Domain Sub-group results of factors contributing to variations in estimated SMD among included studies 1st, first; ROB, Risk of Bias; D, Domain Based on the results of the meta-regression analysis, assessment times of thirst among studies were found to have an insignificant contribution to the variations in the reported SMD for each intervention. The meta-regression results showed that the studies showed less effectiveness with the increase of time between the baseline or first intervention and the last assessment point (see supplementary appendix). However, the results showed no significant correlation between assessment time and intervention effectiveness (P = 0.872). In fact, for every 22.6 h of increase in evaluation time, only around 0.1 changes in effect size will be observed (Coefficient: 0.0000738, (95%CI:−0.00082, 0.00097)). Also, after removing the effect of assessment time from the overall estimate, the heterogeneity did not change considerably. Assessment time didn’t explain variations in the pooled effect size estimates (r2 = 00.00) (see Table 3). Table 3Meta-regression analysis resultsVariablesCoefficient (95% CI)P-valueI2-residualTau2Adj. R2Assessment Times0.0000738 (−0.00082, 0.00097)0.87295.451.040.00 Meta-regression analysis results Given the significant heterogeneity, subgroups were assessed according to the intervention type, surgical procedure, geographic region, total score of bias risk assessment and its domains, fasting duration, mean age, and baseline inequalities. Despite observing a decrease in heterogeneity in certain instances following subgroup analysis, this decrease was considerable in only two sub-groups (a reduction of under 50%) [45]; thus, the starting point for evaluations and baseline inequality could justify the high heterogeneity (Table 2). The results of publication bias assessments via funnel plot showed that studies with smaller sample sizes had higher SMD values. Thus, there seems to be a potential for publication bias (Fig. 5). Despite this, we identified evidence of publication bias for the effects of oral stimulants through Begg’s test for small-study effects (P = 0.009), and the regression-based Egger test (P = 0.0002, beta= −10.72). Funnel plot for assessing potential publication bias Sensitivity analysis using the leave-one-out method showed that omitting each of the studies of Aroni et al., Eren et al., Rajeshkumar et al., Serato et al., and Can et al. [2, 39, 41–43] had the most significant impact on the overall estimated SMD, with a change of over 0.1 (Fig. 6). Another approach in sensitivity analysis by removing the studies with a high risk of bias showed a 0.07 change in pooled SMD (−1.31, (95% CI:−1.87, −0.75)). The 95% CI after removing the mentioned studies had an overlap of above 50% compared with the one that included all studies. In the survey conducted by Mert et al. (2024) [6], two intervention groups were included; however, only one group was necessarily incorporated into the primary meta-analysis. Sensitivity analysis, considering Ice Popsicle as the primary intervention for the Mert et al. (2024) [6] study, did not substantially change the effect size (see Supplementary Appendix). Results of sensitivity analysis via leave one out method All studies were RCTs, and the certainty of evidence was initially assigned a score of 4. The studies were not downgraded due to risk of bias, as more than 50% of them were classified as “low risk” for performance bias and bias related to missing data. However, regarding the homogeneity of the results from the initial studies, considerable heterogeneity was observed among the studies. Only two subgroup analyses justified this heterogeneity. Furthermore, most interventions were supplemented with additional interventions that may have influenced the outcomes. Consequently, a negative score was considered in this context. The outcomes of the funnel plot, as well as Begg and Egger’s tests, indicated potential publication bias. In summary, the quality of evidence provided by the RCTs was downgraded by five levels due to inconsistency, indirectness, and publication bias. Conversely, two levels were upgraded owing to a large effect size, and one level was elevated due to dose-response considerations. Ultimately, the certainty of evidence regarding thirst intensity as the primary outcome was rated as low (see Table 4). Table 4Summary of the findings and the degree of certainty in the supporting evidence. GRADE Working Group grades of evidence. High quality: We have a high level of confidence that the true effect is nearly identical to the estimated effect. Moderate quality: Our level of confidence in the effect estimate is moderate, as the true effect is expected to be in close proximity to the estimated effect. However, there is a potential for significant divergence. Low quality: We have limited confidence in the estimated effect, as the true effect may significantly deviate from the estimate: -1 score, : -2 score, : 0 score, : +1 score, : +2 score Summary of the findings and the degree of certainty in the supporting evidence. GRADE Working Group grades of evidence. High quality: We have a high level of confidence that the true effect is nearly identical to the estimated effect. Moderate quality: Our level of confidence in the effect estimate is moderate, as the true effect is expected to be in close proximity to the estimated effect. However, there is a potential for significant divergence. Low quality: We have limited confidence in the estimated effect, as the true effect may significantly deviate from the estimate : -1 score, : -2 score, : 0 score, : +1 score, : +2 score Cold oral stimulants and thirst discomfort Only three studies evaluated the perioperative thirst discomfort. Meta-analysis showed that the pooled SMD was − 1.24 (95% CI:−1.79, −0.7). Results showed that the interventions significantly lowered the discomfort (p < 0.001). The Q Cochrane heterogeneity test was insignificant, and I2 was estimated to be around 56.9% (see Supplementary Appendix). Results of Egger’s test (P = 0.43) showed that there was no potential for publication bias. Also, the trim and fill method results yielded no additional studies. Due to a lack of data, we couldn’t implement any subgroup analysis. The certainty of the evidence profile is presented in Table 4. Cold oral stimulants and dry mouth Only five studies evaluated the dryness of the mouth. Meta-analysis showed that the pooled standardized mean difference was − 1.44 (95% CI:−3.50, 0.61). Also, we didn’t observe any significant decrease in dry mouth in the intervention group compared to the control group (p = 0.17). The Q Cochrane heterogeneity test was substantial, and I2 was estimated at 98.01% (see Supplementary Appendix). Results of Egger’s test (P < 0.001) showed that there was potential for publication bias. Also, the trim and fill method results didn’t yield additional studies. Due to a lack of data, we couldn’t implement any subgroup analysis. The certainty of the evidence profile is presented in Table 4. Discussion The meta-analysis conducted in our study demonstrated a significant link between the use of oral stimulants in post-surgery patients and a marked decrease in the severity of thirst reported by the intervention group. According to the SMD interpretive regions, the magnitude of this effect size is considered large. The evidence suggests that oral stimulants result in a large reduction in thirst intensity among post-operative patients. Moreover, our analysis revealed that the MD between the intervention and control groups following the intervention, based on the pooled results of 14 studies, was −2.30. This indicates that the intervention group experienced a 2.30-unit greater reduction than the control group. Although no consensus criteria exist for minimally important clinical difference (MCID) in the assessment of thirst intensity, an NRS reduction of 1.7 points has been used as MCID in a previous study focusing on thirst management [46]. The measured MD in our study reached the MCID. Before our study, existing systematic reviews and meta-analyses assessed the intensity of thirst among postoperative patients. However, we were the first to use the GRADE approach to determine the level of certainty. The findings of a recent systematic review and meta-analysis of RCTs examining the effectiveness of menthol in alleviating patient thirst indicated that these interventions are efficacious during pre-and postoperative periods. The study found that the efficacy was greater for preoperative thirst, but the effectiveness did not persist in the postoperative period due to a wide 95% CI [47]. Our study exclusively examined the postoperative phase. The pooled results of six studies in our analysis indicate that menthol demonstrates a high level of effectiveness in reducing the intensity of postoperative thirst among patients. However, the substantial heterogeneity observed across the results may cast doubt on the reliability of this finding. According to recently published studies, mentholated agents exhibited greater efficacy and more consistent results, as narrower 95% CIs indicated. Further primary research is required in this field to establish the effectiveness with a higher level of certainty in the evidence. Past theoretical investigations have consistently demonstrated that menthol activates cold receptors in the oral and nasal cavities. Furthermore, examining the theoretical justification for menthol’s gratifying impact on thirst and respiration and its influence on arousal could clarify the widespread appeal of menthol and the extensive consumption of products containing menthol daily [48, 49]. Generally, it has been demonstrated that menthol modulates oral sensations of warmth and cold [50]. The review studies’ outcomes validate these substances’ application in clinical settings. However, additional research is necessary to establish the certainty of the findings. Another study, which consisted of a systematic review and meta-analysis, assessed the efficacy of cold oral stimuli, as examined in our study. In the study mentioned above, the reviewers utilized the MD as the primary metric, which hinders the comparison of effects across interpretation domains. Nevertheless, the study’s results discussed above indicated that the interventions yielded greater effectiveness compared to routine care [19]. Despite the inclusion of only RCTs in our study, the addition of recently published studies led to results consistent with the findings of Tsai et al.‘s study. In 2024, a systematic review without meta-analysis was conducted to assess the impact of cold oral applications on postoperative thirst management. The interventions were classified into two subgroups: ice and cold water. The results indicated a significant decrease in the intervention group for all studies in the ice application group, including frozen drinking water, frozen saline, ice cubes of cold flavored water, frozen mentholated saline, and mentholated ice. Out of the three studies conducted, two in the cold water group showed a preference for cold water application, while another study favored using citric acid [5]. Although the comparisons weren’t head-to-head, the results were satisfying. Our study demonstrated alignment with the study mentioned concerning several interventions and comparators. Our study assessed the secondary outcomes of thirst discomfort and mouth dryness. The research findings indicated that the interventions significantly reduced thirst-related discomfort, with a substantial impact. The evidence for this effect was of moderate certainty. Oral stimulants probably result in a large reduction in thirst discomfort. Although derived from only three studies, the results are considered satisfactory, and conducting additional studies would increase confidence. Nevertheless, the results regarding dry mouth did not demonstrate a significant decrease in oral dryness. It is worth noting that the estimated effect is very uncertain, and more studies are needed to ascertain the results. Multiple plausible physiological mechanisms may account for the swift diminution of postoperative thirst observed after oral stimulation with cold water, ice, or menthol. Cold and menthol activate oropharyngeal cold-sensitive transient receptor potential melastatin 8 (TRPM8) channels, thereby generating pre-absorptive satiety signals that inhibit the central thirst drive independently of alterations in plasma osmolality [51–53]. In parallel, thermal and chemical stimulation of the oral mucosa increases salivary secretion and improves mucosal wetting, thereby directly reducing the sensation of oral dryness and oropharyngeal discomfort [54, 55]. Menthol additionally may engage trigeminal somatosensory pathways and modulate cortical regions involved in interoception and affective processing (for example, the insula and anterior/mid-cingulate), which may attenuate the affective/emotional component of thirst and amplify perceived relief [56]. Since these mechanisms primarily operate through neural and mucosal pathways rather than through rapid systemic fluid shifts, the effect generally manifests quickly. Still, it is of a transient nature — aligning with clinical trials that demonstrate immediate reductions in subjective thirst, which often diminish in the absence of repeated application or subsequent oral rehydration. Thirst and dry mouth are the predominant symptoms observed in surgical patients. Factors such as the surgical procedure, operation duration, anesthesia type, anesthesia duration, fasting duration, intubation status, and length of stay in PACU influenced patient thirst following surgery [57]. Given the significance of evaluating thirst, multiple instruments exist to measure this parameter. The Analogical Visual Scale (AVE), Verbal Numeric Scale (VNS), Face’s Scale (FS), and Likert-type scales emerged as the most frequently employed measurement tools [58]. The assessment of thirst intensity emerged as the prevailing focus in primary studies. Despite the limited number of studies exploring additional aspects of thirst, we thoroughly assessed these dimensions until we obtained promising results regarding thirst discomfort. This doubles the necessity of future studies on other dimensions. Moreover, all the items mentioned evaluate the subjective perception of thirst, and since they rely on self-reported data, there is a possibility of overestimation [59, 60]. Conducting comprehensive future primary studies examining thirst and dry mouth using objective tools will yield more accurate and realistic outcomes. Ultimately, our study, along with other studies in this field, was conducted to discover optimal solutions to enhance the surgical experience for patients. By incorporating further research in this domain, it is anticipated that more favorable results will be observed in the patient’s clinical setting. Strengths and limitations Our research provides numerous comparative advantages, setting it apart from other studies. Examples of these advantages include conducting a comprehensive search, imposing strict criteria for including RCT studies to ensure a higher level of evidence, and employing the GRADE approach to assess the certainty of evidence. The interpretation of our results requires careful consideration of several limitations. Firstly, a significant number of the studies included in the analysis exhibited a substantial risk of bias. Additionally, some of the studies had co-interventions that could affect the results. Despite strictness in entering studies based on PICO(S) criteria, some included studies didn’t comply with a head-to-head approach in formulating the PICO(S) components. Regarding secondary outcomes, a proper investigation of this clinical heterogeneity among studies was not feasible due to the limited number of studies. It is reasonable to expect that future systematic reviews will encounter fewer challenges due to publishing high-quality studies. As the number of primary studies increases, studies could conduct network meta-analyses to rate the effectiveness of multiple interventions on postoperative thirst management. Furthermore, most studies failed to provide comprehensive details regarding the treatment employed in control groups. It is crucial to accurately document the therapy utilized in the experimental and control groups to facilitate replication of study designs in future research. Furthermore, the evaluations and follow-ups conducted in the studies exhibited such a wide range that it was impossible to find studies that examined identical time periods. A more definitive synthesis of evidence can be achieved through future studies that examine time points with greater stability. Differences were observed in the dosage, treatment interval, and ingredients employed for the stimulation. Nursing research has the potential to establish standardized interventions and create guidelines for clinical healthcare professionals to address postoperative thirst by implementing a secure postoperative protocol. Conclusions According to the results of our study, interventions including menthol, cold water, and ice may result in a large reduction in thirst intensity following surgery. Despite the large magnitude of the effect, the level of evidence remains insufficient, necessitating a doubling of research efforts. Concerning the discomfort caused by thirst, it was also observed that interventions probably result in a large reduction in thirst discomfort following surgery. In terms of dry mouth, no significant efficacy was observed; however, the certainty of this evidence is uncertain, and future research may yield different outcomes. Future research should use uniform methodological strategies to limit methodological heterogeneity and yield more conclusive findings. Supplementary Information Supplementary Material 1. Acknowledgements Our most profound appreciation goes to Rodrigo Assis Neves Dantas, who supplied our research team with the data required for the meta-analysis as the corresponding author of the da Silva et al. article. Our deepest appreciation goes to Rodrigo Assis Neves Dantas, who, in his capacity as the corresponding author of the da Silva et al. article, supplied our research team with the data required for the meta-analysis. Authors’ contributions AZ, SS, and NS were involved in the study’s conception and organization. AN, SB, FM, and HGH were involved in the execution and data collection of the study; AZ, NS, and SS participated in statistical analysis design and execution. AZ, SS, FM, and NS prepared the first draft of the manuscript. All authors contributed to the final manuscript’s review, and all approved the final manuscript. Funding sources No external funding. Data availability The datasets generated and analyzed during the current study are not publicly available but are available from the corresponding author on a reasonable request. Corresponding author`s email for contact: akbar.zarekaseb@gmail.com. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Sogand Sarmadi, Email: sogand.sarmadi@ymail.com. Neda Sanaie, Email: nedasanaie@ymail.com. Akbar Zare-Kaseb, Email: akbar.zarekaseb@gmail.com. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The datasets generated and analyzed during the current study are not publicly available but are available from the corresponding author on a reasonable request. Corresponding author`s email for contact: akbar.zarekaseb@gmail.com.