Association between salivary alkaline phosphatase levels and periodontal disease: a systematic review and meta-analysis Authors: Mario Alberto Alarcón-Sánchez, Sarah Monserrat Lomelí-Martínez, Daniel Ortuño-Sahagún, Julieta Sarai Becerra-Ruiz, Seyed Ali Mosaddad, Artak Heboyan Published: 2026 ABSTRACT AbstractObjectiveThis updated meta-analysis evaluated salivary alkaline phosphatase (ALP) levels in individuals with gingivitis and periodontitis compared to periodontally healthy individuals.Materials and methodsThe review followed PRISMA guidelines and was registered in PROSPERO (CRD420251273461). Searches were conducted in PubMed, Scopus, ScienceDirect, Web of Science, EBSCO, and Google Scholar. Methodological quality was assessed using the JBI Critical Appraisal Checklist for Analytical Cross-Sectional Studies. Meta-analysis was performed using a random-effects model with Review Manager (RevMan) version 5.4.1 and STATA version 19.ResultsTwenty studies (n = 1459) were included; twelve were eligible for meta-analysis. No statistically significant differences emerged in salivary ALP levels between gingivitis and healthy controls (p = 0.15) or between those with gingivitis and periodontitis (p = 0.18). However, individuals with periodontitis had significantly higher salivary ALP levels compared to periodontally healthy individuals (MD = 44.62 [95% CI: 22.55 to 66.69]; p = 0.0010), with high heterogeneity (I2 = 100%). The included studies demonstrated variable methodological quality according to the JBI criteria, and the GRADE certainty was very low.ConclusionsCurrent data suggest an association between elevated salivary ALP levels and periodontitis, although methodological heterogeneity requires cautious interpretation and underscores the need for well-designed, standardized future research.Subject terms: Periodontitis, Gingivitis Introduction Periodontal disease refers to a group of chronic inflammatory disorders that compromise the supporting tissues of the teeth [1]. Its pathogenesis begins with a complex interaction between a dysbiotic dental biofilm and an altered host immune response, influenced by various behavioral, systemic, environmental, and genetic factors [2]. The diagnosis, severity, and extent of periodontal disease are determined by standardized clinical and radiographic parameters; however, these indicators reflect the destruction of dental supporting tissues without indicating the current biological activity of the condition [3, 4]. In recent years, the use of biological markers in dentistry has transformed the way in which immune, inflammatory, and destructive processes associated with oral diseases are evaluated, facilitating the early identification of sites with pathological activity [5, 6]. Among the various available biofluids, saliva has acquired increasing interest because it contains a wide variety of molecules derived from both the host and the microbiota, including proteins, nucleic acids, and various metabolites [7]. Saliva can be obtained quickly, safely, and noninvasively, allowing for repeated sample collection with minimal discomfort to the patient; for this reason, it has been used in research related to the diagnosis and monitoring of periodontal diseases [8]. Furthermore, the presence of enzymes released during the inflammatory response and the destruction of periodontal tissues has driven the use of saliva as a practical alternative for assessing periodontal status and monitoring response to treatment [9]. Among the most studied salivary enzymes is alkaline phosphatase (ALP), a zinc-dependent hydrolase that catalyzes the removal of phosphate groups from various substrates, releasing inorganic phosphate (Pi), an essential component of bone mineralization [10]. Due to its role in the metabolism of mineralized tissue and in bone remodeling, ALP is considered a marker of osteoblastic activity and tissue turnover [11]. In periodontitis, increased ALP activity in saliva reflects enhanced alveolar bone remodeling and the inflammatory response associated with the destruction of periodontal tissues [12], which has led to its study as a potential biomarker of disease activity and severity. A previous meta-analysis confirmed that patients with periodontitis have significantly higher salivary ALP levels than periodontally healthy individuals. However, that evidence was limited to studies published through 2016, analyzed ALP as one of several enzymatic biomarkers, and did not include comparisons with patients with gingivitis; furthermore, there was high heterogeneity among the included studies [13]. These limitations justify conducting an updated systematic review and meta-analysis focused specifically on salivary ALP. Objectives To evaluate ALP levels in saliva samples from subjects with gingivitis and periodontitis compared to periodontally healthy individuals. To evaluate ALP levels in saliva samples from subjects with gingivitis and periodontitis compared to periodontally healthy individuals. To correlate ALP levels with clinical periodontal parameters. To correlate ALP levels with clinical periodontal parameters. Materials and Methods This systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [14] and was registered in the International Prospective Register of Systematic Reviews (PROSPERO - ID: CRD420251273461) on December 25, 2025. The researchers aimed to answer the following research question: Are ALP levels in saliva samples different in subjects with gingivitis and periodontitis compared to periodontally healthy individuals? To address this question, prospective cross-sectional and longitudinal clinical studies were included if they evaluated ALP levels using spectrophotometric, colorimetric, or enzymatic methods in saliva samples from individuals of both sexes, older than 18 years, and diagnosed with gingivitis or periodontitis according to clinical diagnostic criteria such as probing depth (PD), clinical attachment loss (CAL), and/or radiographic bone loss (RBL). Studies that include only patients with systemic diseases (e.g., diabetes mellitus, cardiovascular diseases, metabolic disorders, or autoimmune diseases), those receiving anti-inflammatory drugs, antibiotics, or corticosteroids, individuals with previous periodontal treatment (phase I or II), pregnant or breastfeeding women, and populations composed exclusively of smokers or alcohol consumers were excluded. Reports and case series, randomized controlled trials, in vitro or animal studies, brief communications, conference abstracts, bibliometric studies, systematic reviews, and meta-analyses were also excluded. Information sources and search strategy M.A.A.S. and S.M.L.M. conducted the article search in the following databases: PubMed, Scopus, ScienceDirect, Web of Science (WOS), EBSCO (Dentistry & Oral Science Source), and, as gray literature, Google Scholar. Combinations of the MeSH terms “Periodontal Disease,” “Periodontitis,” “Gingivitis,” “Saliva,” and “Alkaline Phosphatase” were used with the Boolean operators “OR” and “AND.” Language restrictions were applied, and only articles published in English were included. The search was conducted up to November 5, 2025. An updated search for articles was conducted covering the period from January to June 2026. The PubMed search strategy was: ((((((Disease, Periodontal) OR (Diseases, Periodontal)) OR (Periodontal Disease)) OR (Periodontitis)) OR (Gingivitis)) AND (Saliva)) AND (Alkaline Phosphatase), yielding 64 results. For Scopus, the strategy was: TITLE-ABS-KEY (“Disease, Periodontal” OR “Diseases, Periodontal” OR “Periodontal Disease” OR “Periodontitis” OR “Gingivitis”) AND TITLE-ABS-KEY (“Saliva” OR “Salivary”) AND TITLE-ABS-KEY (“Alkaline Phosphatase”) AND NOT TITLE-ABS-KEY (“Review”), yielding 87 results. For ScienceDirect, the strategy was: (“Disease, Periodontal” OR “Diseases, Periodontal” OR “Periodontal Disease” OR “Periodontitis” OR “Gingivitis”) AND (“Saliva”) AND (“Alkaline Phosphatase”) AND (“Nonsurgical Periodontal Therapy”) NOT (“Review”), yielding 1 result. For WOS, it was: (((((ALL = (Periodontal Disease)) OR ALL = (Periodontitis)) OR ALL = (Gingivitis)) AND ALL = (Saliva)) AND ALL = (Alkaline Phosphatase)) NOT ALL = (Review), yielding 38 articles. For EBSCO, it was: ((((((Disease, Periodontal) OR (Diseases, Periodontal)) OR (Periodontal Disease)) OR (Periodontitis)) OR (Gingivitis)) AND (Saliva)) AND (Alkaline Phosphatase) AND (Nonsurgical Periodontal Therapy), yielding 2 articles. For Google Scholar, the search was: Alkaline Phosphatase AND Saliva OR Salivary AND Periodontitis OR Gingivitis OR Periodontal Disease AND Nonsurgical Periodontal Therapy NOT Review NOT Systematic Review NOT Meta-Analysis OR Meta-Analyses, yielding 400 articles. In total, 592 articles were identified. Selection process References were transferred to the Rayyan platform (www.rayyan.ai) [15], accessed on November 10, 2025. Duplicates were removed, and titles and abstracts were reviewed independently. The full texts of the studies were then examined to determine inclusion based on the established eligibility criteria. Two reviewers (M.A.A.S. and J.S.B.R.) independently conducted the selection process. Discrepancies were resolved by consensus among all team members. Data collection process The variables extracted for each study included the author’s name and year of publication, country, study design, journal name, ethics committee approval number, sex, age, number of cases with gingivitis and periodontitis, number of controls (periodontally healthy), total study population, characteristics of saliva collection (method, time, duration, volume, type of saliva, and processing), type of periodontal disease classification used, cut-off values of the clinical parameters considered, the evaluation of other enzymatic markers, biochemical assay or kit used, ALP levels in both study groups, and the correlations obtained. Two reviewers (M.A.A.S and S.M.L.M.) independently extracted the data, and any discrepancies were resolved by consensus of all team members. Outcomes The primary outcomes included the evaluation of ALP levels in saliva samples from subjects with gingivitis and periodontitis compared to periodontally healthy individuals. Secondary outcomes included the analysis of correlations with periodontal clinical parameters. Data items A qualitative and quantitative synthesis was performed based on the collected data. Meta-analyses were conducted using means and standard deviations of ALP levels. Only studies reporting ALP catalytic activity in directly comparable units (U/L or IU/L) were included in the quantitative synthesis (statistical analysis). Quality methodological assessment The methodological quality of all included studies was assessed using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Analytical Cross-Sectional Studies [16]. Two reviewers (M.A.A.S. and J.S.B.R.) conducted this phase independently, and any discrepancies were resolved by consensus among all team members. Effect measures, synthesis methods, and reporting bias assessment Statistical analyses were conducted using Review Manager (RevMan) version 5.4.1 and STATA version 19. Studies reporting ALP activity in directly comparable units (U/L or IU/L) were included in the quantitative synthesis. Because 1 U is equivalent to 1 IU for enzyme activity, no numerical unit conversion was required. Pooled effects were reported as mean differences (MDs) with 95% confidence intervals (CIs). A random-effects model using the inverse-variance method was applied, and CIs were estimated using the Hartung-Knapp-Sidik-Jonkman approach. Statistical heterogeneity was assessed using the I² statistic and the restricted maximum likelihood (REML) estimator for between-study variance (τ²), with I² values > 75% considered indicative of substantial heterogeneity [17]. To complement the pooled effect estimate, 95% prediction intervals were calculated to estimate the expected range of effects in future studies. Sensitivity analysis was performed using a leave-one-out approach, sequentially excluding each study to evaluate the robustness of the pooled estimates. Prespecified subgroup analyses were conducted by saliva type (stimulated vs. unstimulated), study design (cross-sectional vs. prospective longitudinal), and publication period (before 2015 vs. 2015 onwards). Differences between subgroups were evaluated using the χ² test for subgroup differences. Potential small-study effects were assessed by visual inspection of funnel plots and Egger’s regression test, which evaluates funnel plot asymmetry by regressing standardized effect estimates on study precision. Statistical significance was set at p < 0.05. Certainty assessment The certainty of the evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach [18]. Results Study selection A total of 592 articles were identified in the databases consulted, and 80 duplicates were removed, leaving 512 articles. During the screening phase, 469 studies were excluded based on title and abstract, leaving 43 potentially eligible articles. After full-text review, 23 articles were excluded (smokers n = 2; patients with aggressive periodontitis n = 1; retracted article n = 1; postmenopausal women n = 1; patients with diabetes and periodontal disease n = 1; duplicates n = 1; thesis n = 1; no access n = 5; other languages n = 6; gingival crevicular fluid n = 1; abstract n = 1; serum n = 1; review article n = 1). Therefore, 20 articles [11, 19–37] were included in this systematic review, of which 12 were included in the meta-analysis (Fig. 1). PRISMA flow chart diagram. Study characteristics Of the included articles, 7 (35%) were prospective longitudinal clinical studies, and 13 (65%) were cross-sectional studies. Most studies were published after 2012 (14; 70%); the oldest was from 1983 and the most recent from 2024, spanning 41 years. Twenty articles were published in 11 different countries. Seven articles (35%) were published in India, three (15%) in Japan, two (10%) in Iran, and the remainder (5% each) in Iraq, Saudi Arabia, Switzerland, North Macedonia, Nigeria, Romania, Serbia, and the USA (Table 1). Table 1Baseline features of included studies in this systematic review.Author’s and yearCountryStudy TypeJournalEthics Approval (number)SexMa/FeAverage agen (CG/GG/PG)n (Total)/SamplesSalivary featuresTime/Amount /TypeCollection /ScheduleProcessingKido et al. [12]JapanCross-sectionalBMC Oral HealthYes, Tokushima University Hospital Ethics Committee (No. 4225)37/5658.926/NR/679310 min1–5 mLUnstimulated salivaPassive droolingNRCentrifuged at 3000 × g for 10 min at 4 °CAl-Maliki et al. [19]IraqCross-sectionalJ. Med Pharm. Chem. Res.Yes, number not specifiedNR18–6025/NR/255060 seconds1.5 mlStimulated salivaNRParaffinNRCentrifuged at 15,000 g for 10 minutes at 4 °C. Supernatant stored at −20 °CParihar et al. [20]IndiaProspective longitudinalNatl J Maxillofac SurgYes, Dean/2021/EC/2722NR25–5536/NR/3672NR5 mlUnstimulatedsalivaNR9:00–11:00 amCentrifuged at 3000 rpm for 5 minutes at 4 °C. Supernatant stored at −80 °CRasaei et al. [21]IranCross-sectionalJ Fam Med Prim CareNRNRNR23/NR/23462 min3 mlUnstimulated salivaNRNRCentrifuged at 4000– 4500 rpm for 3–5 minutes. NRKoppolu et al. [22]Saudi ArabiaProspective longitudinalSaudi J Biol SciNR75/6020–5539/48/48135NRNRUnstimulatedsalivaNRNRCentrifuged at 3000 rpm for 10 minutes. NRRamenzoni et al. [23]SwitzerlandCross-sectionalDiagnosticsBASEC-Nr. 2018-0022124/1635–6020/NR/204015 min/5 minNR Unstimulated and stimulated salivaExpectorationParafilmNRCentrifuged 6000 rpm × 10 min at 4 °C. Supernatant stored at −80 °CRisteska et al. [24]North MacedoniaCross-sectionalContributions, Sec. of Med. Sci.NR23/2219–7015/15/1545NRNRUnstimulatedsalivaSpittingMorning collectionSupernatant stored at −20 °C. NRYaghini et al. [25]IranProspective longitudinalDent Implant Periodontol Adv JYes, number not specifiedNR20–55NR/NR/16165 minNRUnstimulatedsalivaNRMorning collectionCentrifuged at 3500 rpm. NRJeyasree et al. [26]IndiaProspective longitudinalJ Indian Soc PeriodontolIRB approval No. 2014-MD-BrII-SAS-05NR30–5010/NR/4050NR5 mlUnstimulatedsalivaSpitting9:00–11:00 amCentrifuged 3000 rpm × 5 min. NRPatel et al. [27]IndiaCross-sectionalJ Clin Diagn ResKIMSDU/IEC/01/2013 (06/12/2013)60/9020–4550/50/50150NR5 mlUnstimulatedsalivaNR10:00–11:00 amTransported within 24 h at 2–4 °C. NRDeepika et al. [28]IndiaCross-sectionalJ Pharm Sci & ResNRNR25-6015/NR/1530NRNRUnstimulatedsalivaNRNRNRLuke et al. [29]IndiaCross-sectionalJ Int Oral HealthYes, number not specified40/021–5010/15/1540NR5 mlUnstimulatedsalivaDraining methodNRCentrifuged 10,000 rpm × 10 min. NRDabra et al. [30]IndiaProspective longitudinalJ Indian Soc PeriodontolNRNRNR20/20/20605 min10 mlUnstimulatedsalivaParaffinNRCentrifuged 10,000 rpm × 10 min. NRNomura et al. [31]JapanProspective longitudinalArch Oral BiolApproved by IRBs of participating universities (numbers NR)22/6360.16NR/NR/85855 minNRUnstimulatedsalivaParaffinNRMaintained at 4 °C and sent immediately for enzyme analysisIkekpeazu et al. [32]NigeriaCross-sectionalInt J Diabetes Dev CtriesApproved by institutional ethics committee (number NR)Unclear28–8180/NR/80160NR5 mlUnstimulatedsalivaNRNRCentrifuged 3000 rpm × 10 min; supernatant analyzed immediately; NRAgnihotram et al. [33]IndiaCross-sectionalInt J Appl Biol Pharm TechnolYes, Declaration of Helsinki; written informed consent (ethics number NR)NR30–6526/0/5480NRNRUnstimulated salivaAutomatic pipette (Salivette®)Centrifuged at 10,000 rpm for 10 min, NRTotan et al. [34]RomaniaCross-sectionalClin Chem Lab MedNR28/2225–5525/NR/2550NR1 mlUnstimulatedsalivaNR9:00–11:00Centrifuged 500 g × 10 min, NRNomura et al. [35]JapanCross-sectionalJ Oral Sci.Yes, Ethics Committee of Nippon Dental College (ethics number NR)137/5037.221–6063/23/1011875 minNRStimulated salivaChewing tasteless/odorless gum baseNRSamples were analyzed the same day (NR). ALP was measured with L-type Wako ALP J commercial kitTodorovic et al. [36]SerbiaProspective longitudinalMed Oral Patol Oral Cir BucalNRNR25–5020/NR/3050NRNRUnstimulatedsalivaNRNRCentrifuged at 10,000 rpm x 10 min; analyzed immediately (No storage)Nakamura et al. [37]USACross-sectionalJ Periodontal ResNRNRNR10/NR/1020NR8 mlUnstimulatedsalivaNRNRCentrifuged 5000 × g for 15 min at 4 °C; supernatant/sediment assayed immediately or stored at −70 °CNR Not reported, Ma Male, Fe Female, CG Control group, GG Gingivitis group, PG Periodontitis Group, Age is represented as mean or range. Baseline features of included studies in this systematic review. 10 min 1–5 mL Unstimulated saliva Passive drooling NR 60 seconds 1.5 ml Stimulated saliva NR Paraffin NR NR 5 ml Unstimulated saliva NR 9:00–11:00 am 2 min 3 ml Unstimulated saliva NR NR NR NR Unstimulated saliva NR NR 15 min/5 min NR Unstimulated and stimulated saliva Expectoration Parafilm NR NR NR Unstimulated saliva Spitting Morning collection 5 min NR Unstimulated saliva NR Morning collection NR 5 ml Unstimulated saliva Spitting 9:00–11:00 am NR 5 ml Unstimulated saliva NR 10:00–11:00 am NR NR Unstimulated saliva NR NR NR 5 ml Unstimulated saliva Draining method NR 5 min 10 ml Unstimulated saliva Paraffin NR 5 min NR Unstimulated saliva Paraffin NR NR 5 ml Unstimulated saliva NR NR NR NR Unstimulated saliva NR 1 ml Unstimulated saliva NR 9:00–11:00 37.2 21–60 5 min NR Stimulated saliva Chewing tasteless/odorless gum base NR NR NR Unstimulated saliva NR NR NR 8 ml Unstimulated saliva NR NR NR Not reported, Ma Male, Fe Female, CG Control group, GG Gingivitis group, PG Periodontitis Group, Age is represented as mean or range. A total of 1459 participants were included: 171 had gingivitis, 775 had periodontitis, and 513 were periodontally healthy. The age range was 18 to 81 years. Among the participants, 30.5% were men, 25.9% were women, and the sex of the remaining 43.6% was not reported. For saliva sample processing, three studies analyzed paraffin-stimulated saliva, while the others analyzed unstimulated saliva obtained by expectoration, spitting, or drainage. Collection times ranged from 1 to 15 minutes. On average, 5 mL of unstimulated saliva and 1.5 mL of stimulated saliva were collected. Collection occurred in the morning between 9:00 and 11:00 a.m. Saliva samples were centrifuged for 3 to 10 minutes and stored at −20 °C to −80 °C. The studies used spectrophotometric, colorimetric, and enzymatic methods to determine salivary ALP levels, which were reported in various units, including IU/L, U/L, mU/mL, mg/dL, IU, and µg/dL. Other molecules analyzed included lactate dehydrogenase, alanine aminotransferase, arginase, acid phosphatase, lactoferrin, calprotectin, calcium, phosphate, pH, osmolarity, viscosity, density, leukocytes, hemoglobin, aspartate aminotransferase, creatine kinase, gamma glutamyl transferase, plasma homocysteine, C-reactive protein, total cholesterol, triglycerides, low-density lipoprotein, high-density lipoprotein, and blood urea nitrogen (Table 2). Table 2Characteristics of ALP in saliva of individuals with periodontal disease and periodontally healthy individuals.Author’s and YearClinical parameters assessmentOther biomarkers assessmentTechniqueUnits of measurementValues GC (ALP)Values GG/PG (ALP)P-valuesOutcomes (levels and correlations)Kido et al. [12]Stage II–IV periodontitis (2018 classification); plaque index, probing depth, clinical attachment level, bleeding on probing, gingival index, periodontal inflamed surface area and alveolar bone loss evaluatedCalprotectin, hemoglobin, lactate dehydrogenase, alanine aminotransferase, aspartate aminotransferaseEnzymatic colorimetric spectrophotometric method (enzyme kinetics); automated biochemistry analyzer (BioMajesty™ JCA-BM6070/C); ALP-IFCC kit (Ekdia® XL)U/L6.4 ± 5.5NR/17.5 ± 14.5<0.001↑ Salivary ALP in periodontitis compared with controls. ALP positively correlated with clinical indicatorsAl-Maliki et al. [19]UnclearLactate dehydrogenase, arginaseEnzymatic colorimetric spectrophotometric method/IFFCC mod. Kits®UI/L21.88 ± 1.7953.36 ± 4.24<0.05↑ ALP in periodontitis group compared control groupParihar et al. [20]UnclearNot reportedEnzymatic colorimetric spectrophotometric method (Autospan reagents)U/L23.44 ± 4.76105.66 ± 16.33Postoperative: 49.54 ± 5.69 (1 month post-therapy)<0.001↑ Salivary ALP in periodontitis group; significant reduction after NSPTRasaei et al. [21]Probing depth: 5–7 mmNot reportedEnzymatic colorimetric spectrophotometric method/ALP kit (Pars Azmoun) + Hitachi autoanalyzerU/L24.78 ± 4.3780.17 ± 23.9<0.001↑ ALP in periodontitis group compared control groupKoppolu et al. [22]Probing depth: ≥4 mmAcid phosphataseEnzymatic colorimetric spectrophotometric method/autoanalyzer (Autospan reagents)U/L43.56 ± 10.8752.06 ± 30.47123.54 ± 83.90Postoperative:41.03 ± 24.7757.24 ± 36.28 (1 month post-therapy)<0.05↑ Salivary ALP in gingivitis and periodontitis groups compared control group; significant reduction after NSPT; positive correlation with clinical parametersRamenzoni et al. [23]Stage III grade B generalized periodontitis (2018 classification): interdental clinical attachment loss: ≥5 mm, radiographic bone loss to middle third, probing depth: ≥6 mmLactoferrin, calcium, phosphate, pH, osmolarity, viscosity, density, leukocytes, hemoglobinEnzymatic colorimetric spectrophotometric method/Kinetic colorimetric ALP assay (Abcam)mU/mL0.97 (0.68) (median, IQR; unstimulated) 0.64 (0.95) (median, IQR; stimulated)2.46 (3.94) (median, IQR; unstimulated) 0.90 (2.21) (median, IQR; stimulated)0.030.2↑ Salivary ALP in periodontitis group compared control group (unstimulated saliva only)Risteska et al. [24]Clinically manifest periodontitis: clinical attachment loss: ≥4 mm on ≥2 surfaces of ≥2 teethAspartate aminotransferase, lactate dehydrogenaseEnzymatic colorimetric spectrophotometric method/(ALP-IFCC)U/L10.43321.46644.893<0.000↑ Salivary ALP in periodontitis group compared control groupYaghini et al. [25]Chronic periodontitis (AAP criteria): interdental clinical attachment loss: > 4 mm at ≥2 non-adjacent teeth or probing depth: > 5 mm at ≥2 interdental sites; probing depth categorized as mild (1–2 mm), moderate (3–4 mm), severe (>5 mm)Calcium, phosphorusEnzymatic colorimetric spectrophotometric method/(Autoanalyzer BT-3000; Pars Azmoon kit)mg/dLNot reported48.31 Postoperative: 40.37 (1 month post-therapy)0.306No significant change in salivary ALP after NSPTJeyasree et al. [26]Chronic generalized periodontitis: ≥5 qualifying sites in ≥2 quadrants, ≥2 affected teeth/quadrant, probing depth: ≥5 mm, clinical attachment loss: ≥3 mm, bleeding on probing: presentNot reportedEnzymatic colorimetric spectrophotometric method (semi-autoanalyzer; Diasys® kit; IFCC method)UI23.00 ± 6.6779.55 ± 6.40 Postoperative: 49.47 ± 5.11 (1 month post-therapy)<0.001↑ Salivary ALP in periodontitis group; significant reduction after NSPTPatel et al. [27]Chronic generalized periodontitis (1999 classification): periodontal pockets >4 mm with clinical attachment loss (with/without bleeding on probing)Calcium, phosphorous, pHEnzymatic colorimetric spectrophotometric method (Accurex)µg/dL20.70 ± 7.1833.06 ± 5.4949.62 ± 16.46<0.001↑ Salivary ALP in gingivitis and periodontitis groups compared control groupDeepika et al. [28]Periodontitis diagnosed by clinical parameters: gingival index, bleeding on probing, probing depth (radiographic criteria not reported)Aspartate aminotransferase, creatine kinaseEnzymatic colorimetric spectrophotometric method (IFCC-recommended for ALP)Not reportedNot reportedNot reportedNot reported↑ Salivary ALP in periodontitis group compared control groupLuke et al. [29]Chronic periodontitis: diagnosis based on OHI-S, SBI, Russell’s periodontal index, probing depth, clinical attachment lossAspartate aminotransferase, alanine aminotransferase, blood urea nitrogenEnzymatic colorimetric spectrophotometric method (Enzopak kinetic kit)IU/L33.48 ± 15.8476.85 ± 36.77164.39 ± 54.45<0.001↑ Salivary ALP in gingivitis and periodontitis groups compared control groupDabra et al. [30]Periodontal disease classified by gingival index (Löe & Silness), probing, and clinical attachment loss; periodontitis: ≥4 teeth with clinical attachment loss: ≥5 mm and moderate radiographic bone lossAspartate aminotransferase, alanine aminotransferase, gama glutamyl transferase, acidic phosphataseEnzymatic colorimetric spectrophotometric method (Hitachi autoanalyzer)U/L24.85 ± 12.1443.20 ± 16.5744.55 ± 19.14 Postoperative:28.53 ± 11.0329.43 ± 13.10 (3 weeks post-therapy)<0.001↑ Salivary ALP in gingivitis and periodontitis groups compared control group; significant reduction after NSPTNomura et al. [31]Chronic periodontitis diagnosed by probing depth, clinical attachment loss, bleeding on probing, gingival; progression defined as clinical attachment loss: > 3 mm at ≥1 site over 18 monthsAspartate aminotransferase, alanine aminotransferase, lactato dehydrogenase, free hemoglobin; periodontal pathogensEnzymatic colorimetric spectrophotometric method (Wako ALP kit)U/LNot reported13.3 ± 14.2 (non-progressive CP); 21.6 ± 31.2 (progressive CP)0.23ALP higher in progressive periodontitis, but not statistically significantIkekpeazu et al. [32]Periodontal disease: inflammation and/or bleeding on probing and presence of hard depositsAcid phosphatase, aspartate transaminase, alanine transaminase, creatine kinase, lactate dehydrogenaseEnzymatic colorimetric spectrophotometric methodU/L7.40 ± 0.4436.30 ± 1.44<0.001↑ Salivary ALP in periodontal disease vs healthy controlsAgnihotram et al. [33]Chronic periodontitis diagnosed based on the presence of calculus and plaque, ≥1 periodontal pocket >6 mm in each quadrant, alveolar bone destruction, gingival index, bleeding on probing, and probing depthCreatine kinase, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, acid phosphatase, γ-glutamyl transferase, plasma homocysteine, C-reactive protein, total cholesterol, triglycerides, HDL, LDL, neutrophils, WBC, RBC, thrombocytes, and hemoglobinEnzymatic colorimetric spectrophotometric method (IFCC) using a Hitachi 704 automated analyzerU/L8.20 ± 1.04NR/56.34 ± 23.60<0.01↑ Salivary ALP in chronic periodontitis compared with healthy controls; ALP showed a positive correlation with gingival index and periodontal pocket depth (r = 0.782, P < 0.01)Totan et al. [34]Periodontitis: mean probing ≥5 mm, bleeding on probing present, mean alveolar bone loss ≥40% (radiographic)Aspartate aminotransferase, alanine aminotransferaseEnzymatic colorimetric spectrophotometric method (DiaSys kits; automated analyzer)U/L6.6 ± 4.234.38 ± 7.5<0.001↑ Salivary ALP in periodontitis group compared control groupNomura et al. [35]Periodontal disease assessed using the Community Periodontal Index with a WHO probe. Diagnostic criteria: gingivitis (bleeding on probing), moderate periodontitis ( ≥ 1 site with probing depth 4–5 mm), and severe periodontitis ( ≥ 1 site with probing depth ≥6 mm).Aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, creatinine, blood urea nitrogen, urea, free hemoglobin (f-Hb), and LDH isoenzymes (LDH1–LDH5, H and M subunits).Enzymatic colorimetric spectrophotometric method using commercial clinical chemistry kits (L type Wako ALP J for ALP).IU/L11.0 ± 5.714.1 ± 9.8/17.5 ± 26.30.04Salivary ALP differed significantly among periodontal status groups by ANOVA; however, after Bonferroni correction, ALP was not among the biomarkers that remained statistically significantTodorovic et al. [36]Periodontal disease diagnosed by clinical parameters: gingival index, bleeding on probing, probing depthCreatine kinase, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, gamma glutamyl transferase, acidic phosphataseEnzymatic colorimetric spectrophotometric method (Hitachi 911 analyzer)U/L7.30 ± 2.0538.40 ± 9.89 Postoperative:25.12 ± 7.34 (non-specific)<0.001↑ Salivary ALP in periodontitis group compared control group; significant reduction after NSPTNakamura et al. [37]Adult periodontitis: probing depth: >5 mm on ≥10 teeth; controls periodontally healthyMultiple salivary enzymes (API ZYM panel)Enzymatic colorimetric spectrophotometric method/API ZYM semi-quantitative micromethodnmol liberated chromogen25.5 ± 11.730.0 ± 8.2<0.05↑ Salivary ALP in periodontitis group compared control groupNR Not reported, ALP Alkaline phosphatase, CG Control group, GG Gingivitis group, PG Periodontitis group, IQR Interquartile range, IFCC International Federation of Clinical Chemistry and Laboratory Medicine, NSPT Nonsurgical periodontal therapy. The values were reported as mean ± standard deviation. Characteristics of ALP in saliva of individuals with periodontal disease and periodontally healthy individuals. 105.66 ± 16.33 Postoperative: 49.54 ± 5.69 (1 month post-therapy) 52.06 ± 30.47 123.54 ± 83.90 Postoperative: 41.03 ± 24.77 57.24 ± 36.28 (1 month post-therapy) Enzymatic colorimetric spectrophotometric method/ Kinetic colorimetric ALP assay (Abcam) 0.03 0.2 Enzymatic colorimetric spectrophotometric method/ (ALP-IFCC) 21.466 44.893 33.06 ± 5.49 49.62 ± 16.46 76.85 ± 36.77 164.39 ± 54.45 43.20 ± 16.57 44.55 ± 19.14 Postoperative: 28.53 ± 11.03 29.43 ± 13.10 (3 weeks post-therapy) 38.40 ± 9.89 Postoperative: 25.12 ± 7.34 (non-specific) NR Not reported, ALP Alkaline phosphatase, CG Control group, GG Gingivitis group, PG Periodontitis group, IQR Interquartile range, IFCC International Federation of Clinical Chemistry and Laboratory Medicine, NSPT Nonsurgical periodontal therapy. The values were reported as mean ± standard deviation. Meta-analysis There were no statistically significant differences in ALP levels between patients with gingivitis and the control group (MD = 16.05 [95% CI = −10.14 to 42.25]; p = 0.15; I2 = 91%) (Fig. 2, A), or between patients with gingivitis and those with periodontitis (MD = 38.94 [95% CI = −32.36 to 110.25]; p = 0.18; I2 = 97%) (Fig. 2, panel B). However, a significant increase in ALP levels was observed in patients with periodontitis compared to healthy controls (MD = 44.62 [95% CI: 22.54 to 66.69]; p = 0.0010; I2 = 100%) (Fig. 2, panel C). The 95% prediction interval ranged from −30.40 to 118.54. Comparison of salivary alkaline phosphatase levels according to periodontal status. Forest plots showing the comparison of salivary alkaline phosphatase (ALP) levels between (A) subjects with gingivitis and controls, (B) subjects with periodontitis and gingivitis, and (C) subjects with periodontitis and controls. Heterogeneity assessment Leave-one-out sensitivity analysis Leave-one-out sensitivity analysis showed that sequential exclusion of individual studies did not alter the pooled effect estimate (MD range: 37.67–48.07), with all analyses remaining statistically significant (p ≤ 0.001). Heterogeneity remained high (I² = 99–100%), indicating that no single study disproportionately influenced the overall findings (Supplementary Table 1). Exploratory subgroups analysis To explore potential sources of heterogeneity, subgroup analyses were performed according to saliva type, study design, and publication period (Supplementary Figs. 1–3). When studies were stratified by saliva type, the pooled MD was 19.43 (95% CI: -139.17 to 178.03) for stimulated saliva and 49.94 (95% CI: 24.44 to 75.94) for unstimulated saliva. Considerable heterogeneity persisted within both subgroups (stimulated: I² = 96%; unstimulated: I² = 100%), and no statistically significant subgroup difference was observed (χ² = 3.29, p = 0.07). Similarly, stratification by study design demonstrated significantly higher salivary ALP levels in periodontitis patients in both cross-sectional (MD = 40.88, 95% CI: 9.53 to 72.22; I² = 100%) and prospective longitudinal studies (MD = 52.55, 95% CI: 0.82 to 104.28; I² = 99%). The between-subgroup difference was not statistically significant (χ² = 0.31, p = 0.58). Likewise, in the subgroup analysis according to publication period (before 2015 vs. 2015 onwards), the pooled MD was 27.39 (95% CI: 13.21 to 41.58) for studies published before 2015 and 63.65 (95% CI: 19.74 to 107.57) for studies published from 2015 onwards. Considerable heterogeneity persisted within both subgroups (before 2015: I² = 98%; 2015 onwards: I² = 100%), and a statistically significant subgroup difference was observed (χ² = 4.08, p = 0.04). Overall, considerable heterogeneity persisted across the subgroup analyses. While no significant subgroup differences were observed according to saliva type or study design, a significant difference was observed between publication-period subgroups. Correlation of salivary ALP levels with clinical periodontal parameters Kido et al., 2026, Luke et al., 2015, and Todorovic et al., 2005 demonstrated that salivary ALP levels were positively correlated with PD, CAL, bleeding on probing (BOP), plaque index (PI), gingival index (GI), and RBL [11, 29, 36]. Methodological quality of the included studies Although 7 of the 20 studies included prospective follow-up, only baseline observational comparisons between healthy, gingivitis, and periodontitis groups contributed to the qualitative and quantitative syntheses. Therefore, a single appraisal tool was applied consistently across all studies. Eight studies met all the criteria of this tool (8/8). Five studies met 7/8 criteria, three met 6/8 criteria, two met 5/8 criteria, one met 4/8 criteria, and another met 3/8 criteria. The main methodological deficiencies were related to clearly defined inclusion criteria, diagnosis of the condition using objective and standardized criteria, and strategies for controlling confounding factors (Table 3). Table 3Quality assessment with JBI.StudyQ1Q2Q3Q4Q5Q6Q7Q8JBI criteria metKido et al. [12]YesYesYesYesYesYesYesYes8/8Al-Maliki et al. [19]YesYesYesUnclearYesNoYesYes6/8Parihar et al. [20]YesYesYesYesYesYesYesYes8/8Rasaei et al. [21]YesYesYesYesYesUnclearYesYes7/8Koppolu et al. [22]YesYesYesYesYesYesYesYes8/8Ramenzoni et al. [23]YesYesYesYesYesUnclearYesYes7/8Risteska et al. [24]UnclearYesYesUnclearYesUnclearYesYes5/8Yaghini et al. [25]YesYesYesYesYesYesYesYes8/8Jeyasree et al. [26]YesYesYesYesYesYesYesYes8/8Patel et al. [27]YesYesYesYesYesUnclearYesYes7/8Deepika et al. [28]UnclearYesYesUnclearYesUnclearYesUnclear4/8Luke et al. [29]UnclearYesYesUnclearYesUnclearYesYes5/8Dabra et al. [30]YesYesYesYesYesYesYesYes8/8Nomura et al. [31]YesYesYesYesYesYesYesYes8/8Ikekpeazu et al. [32]YesYesYesYesYesUnclearYesYes7/8Agnihotram et al. [33]YesYesYesYesNoNoYesYes6/8Totan et al. [34]UnclearYesYesYesYesUnclearYesYes6/8Nomura et al. [35]YesYesYesYesYesNoYesYes7/8Todorovic et al. [36]YesYesYesYesYesYesYesYes8/8Nakamura et al. [37]UnclearYesUnclearUnclearUnclearUnclearYesYes3/8Questions:Q1: Inclusion.Q2: Subjects/Environment.Q3: Valid exposure.Q4: Standard Criteria.Q5: Confounders identifiedQ6: Confounding strategies.Q7: Valid outcome.Q8: Adequate statistics. Quality assessment with JBI. Questions: Q1: Inclusion. Q2: Subjects/Environment. Q3: Valid exposure. Q4: Standard Criteria. Q5: Confounders identified Q6: Confounding strategies. Q7: Valid outcome. Q8: Adequate statistics. Publication bias assessment Visual inspection of the funnel plot showed an asymmetric distribution of studies around the pooled effect estimate (Fig. 3). Consistent with this, Egger’s regression test indicated possible small-study effects (β = 5.49, SE = 1.56, z = 3.50, p = 0.0005). Given the limited number of studies and the substantial between-study heterogeneity, these findings should be interpreted with caution and should not be attributed solely to publication bias. Assessment of publication bias in studies comparing salivary alkaline phosphatase levels. Funnel plot of studies included in the meta-analysis comparing salivary alkaline phosphatase (ALP) levels between subjects with periodontitis and periodontally healthy controls. Each point represents an individual study included in the meta-analysis. Each point represents an individual study. The vertical red line indicates the pooled mean difference, and the diagonal lines represent the pseudo 95% confidence limits used to visually assess funnel plot asymmetry. Certainty of evidence results According to the GRADE approach, the certainty of evidence for salivary ALP as a biomarker in periodontal disease was rated as very low ⨁◯◯◯. This rating was mainly due to methodological limitations in the included observational studies, substantial heterogeneity among studies, and imprecision related to small sample sizes. No serious concerns regarding indirectness were identified. Discussion This systematic review and meta-analysis evaluated studies on salivary ALP concentrations in participants with gingivitis and periodontitis, compared to periodontally healthy subjects. The evidence shows that salivary ALP levels are significantly higher in periodontitis than in healthy controls. This increase occurs mainly in the irreversible destructive stages of the disease, rather than in gingival inflammatory mechanisms. The ALP enzyme, expressed in various periodontal cells such as fibroblasts, osteoblasts, and leukocytes, regulates bone homeostasis (formation of hard tissue) [19, 27]. In periodontal disease, microbial imbalance and activation of inflammatory pathways by the host immune system disrupt physiological mechanisms [38, 39]. Persistent inflammatory mediators cause overexpression of cytokines, prostaglandins, and matrix metalloproteinases, which stimulate damage to periodontal tissues, resorption, and bone turnover [38]. As a result, ALP levels in saliva and crevicular fluid increase, supporting its potential as a biomarker of periodontitis activity and severity [11, 19–37]. Multiple studies have documented significantly elevated ALP levels in patients with periodontitis compared to periodontally healthy participants, and these levels have been linked to increased periodontal parameters such as PD, CAL, BOP, PI, GI, and RBL, which reflect the dynamics of bone tissue remodeling [11, 29, 36]. On the other hand, in gingivitis – a condition characterized by a reversible chronic inflammatory process without destruction of periodontal tissues – ALP expression is minimal or insignificant, but higher than in healthy periodontium [22, 24, 27, 29, 30, 35]. Nevertheless, current evidence does not support using it to distinguish gingivitis from periodontitis or to assess disease severity or progression. In the oral cavity, ALP can be identified in various biological fluids. In most studies analyzed, this biomarker was measured in saliva [11, 19–37]. It was also identified in gingival crevicular fluid [20] and serum [20, 22, 26]. Saliva has become a valuable source of information for understanding the local state of the periodontium, as it allows observation of the immunoinflammatory response to tissue destruction and remodeling, providing evidence of biological activity in the context of a pathological process [40, 41]. Saliva also offers several important advantages, including noninvasive, safe, painless collection, easy handling, and low cost without causing tissue damage [6]. Consequently, saliva is a practical, noninvasive biological fluid for measuring ALP and investigating its association with periodontal disease. However, further studies on diagnostic accuracy and longitudinal performance are needed before its clinical utility can be fully established. Although the present meta-analysis included only participants without systemic conditions or other major confounding factors whenever possible, several primary studies explored the behavior of salivary ALP under specific clinical scenarios that provide additional biological context. Ikekpeazu et al. (2011) reported significantly higher ALP levels in patients with both periodontitis and diabetes mellitus than in individuals with periodontitis alone, suggesting that metabolic dysregulation may further enhance tissue destruction and bone turnover [32]. Similarly, Al-Maliki et al. (2024) found elevated salivary ALP concentrations in both smokers and non-smokers with periodontitis compared with healthy controls, indicating that ALP remains increased despite the immunomodulatory effects of tobacco use [19]. In addition, Koppolu et al. (2021) and Jeyasree et al. (2018) observed a significant reduction in salivary ALP levels following non-surgical periodontal therapy, supporting the potential utility of this biomarker for monitoring treatment response in addition to disease detection. Although these findings were not incorporated into the quantitative synthesis because of the predefined eligibility criteria, they reinforce the biological relevance of salivary ALP as a dynamic marker that reflects periodontal inflammatory activity and tissue remodeling under different clinical conditions [22, 26]. This systematic review and meta-analysis has several limitations. First, considerable heterogeneity was observed among the studies and persisted after sensitivity and subgroup analyses, suggesting the potential influence of additional clinical and methodological factors. Although a statistically significant difference was observed between studies published before 2015 and those published from 2015 onwards, considerable heterogeneity remained within both subgroups. Therefore, this subgroup finding should be interpreted cautiously. Second, there was a significant lack of standardization in the protocols for collecting, processing, and analyzing saliva samples, including differences in saliva type, collection time, storage conditions, quantification methods, and units of measurement, which hinders direct comparison between studies. Furthermore, most included studies had methodological limitations in participant selection, diagnostic standardization, and control of confounding factors. Combined with substantial heterogeneity across studies, these limitations resulted in very low certainty of evidence according to GRADE criteria. Finally, the small number of studies evaluating patients with gingivitis, post-periodontal-treatment follow-up, and correlations with clinical parameters limited the ability to conduct more robust quantitative analyses in these settings. Prospective longitudinal studies with standardized methodologies for the collection, processing, and quantification of salivary ALP are needed to improve comparability across studies. Furthermore, it is necessary to establish reference values and diagnostic cut-off points to determine the sensitivity, specificity, and predictive ability of this biomarker at different stages of periodontitis according to the 2018 classification. Future research should also evaluate the utility of ALP in combination with other salivary biomarkers to develop more accurate diagnostic models. Finally, basic studies are needed to analyze the expression and tissue localization of ALP using immunohistochemistry, molecular techniques, and experimental models, as well as research to differentiate the various ALP isoenzymes, with the aim of better understanding their cellular origin and their role in periodontal inflammation and bone remodeling. Conclusion This PRISMA-compliant meta-analysis confirms that salivary ALP levels are significantly higher in individuals with periodontitis than in periodontally healthy controls, reflecting increased bone turnover and inflammation-driven tissue destruction mediated through pathways involving cytokines, prostaglandins, and matrix metalloproteinases. No significant differences were observed between individuals with gingivitis and those with periodontitis. Given its noninvasive nature, ease of collection, low cost, and ability to reflect local immunoinflammatory responses, salivary ALP is a promising biomarker for periodontal disease. However, substantial interstudy heterogeneity, nonstandardized saliva collection and analytical protocols, and the low certainty of the available evidence preclude firm conclusions about its diagnostic accuracy, prognostic value, or clinical utility for disease monitoring. Well-designed longitudinal and diagnostic accuracy studies are needed to establish its potential clinical applications. Supplementary information Supplementary Fig. 3 Supplementary Table 1 Supplementary Fig. 1 Supplementary Fig. 2 Description of supplementary figure Acknowledgements None. Author contributions Mario Alberto Alarcón-Sánchez: Writing—review and editing, Writing—original draft preparation, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Sarah Monserrat Lomelí-Martínez: Writing—review and editing, Writing—original draft preparation, Visualization, Validation, Supervision, Methodology, Formal analysis, Conceptualization. Daniel Ortuño-Sahagún: Writing—review and editing, Writing—original draft preparation, Visualization, Validation, Supervision, Formal analysis. Julieta Sarai Becerra-Ruiz: Writing—review and editing, Writing—original draft preparation, Visualization, Validation, Supervision, Methodology, Formal analysis, Conceptualization. Seyed Ali Mosaddad: Writing—review and editing, Writing—original draft preparation, Visualization, Validation, Supervision, Resources, Project administration, Investigation, Formal analysis, Data curation, Conceptualization. Artak Heboyan: Writing—review and editing, Writing—original draft preparation, Visualization, Validation, Supervision, Resources, Project administration, Investigation, Formal analysis, Data curation, Conceptualization. Funding No external funding was received. Data availability The data supporting this study’s findings are available from the corresponding author upon reasonable request. Competing interests The authors declare no competing interests. Ethical approval Not applicable. Informed consent Not applicable. Consent for publication Not applicable. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Mario Alberto Alarcón-Sánchez, Email: marioaasanchez@hotmail.com. Seyed Ali Mosaddad, Email: Mosaddad.sa@gmail.com. Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41405-026-00488-z. References Associated Data Supplementary Materials Supplementary Fig. 3 Supplementary Table 1 Supplementary Fig. 1 Supplementary Fig. 2 Description of supplementary figure Data Availability Statement The data supporting this study’s findings are available from the corresponding author upon reasonable request.