Risk factors and clinical outcomes of oral mucositis in patients undergoing hematopoietic stem cell transplantation: a retrospective study of 551 cases

Abstract

Background

Oral mucositis (OM) is a common complication in hematopoietic stem cell transplantation (HSCT). This study aimed to analyze OM characteristics, risk factors, and preventive strategies.

Methods

Medical records of 551 patients undergoing HSCT from June 2019 to June 2022 were retrospectively reviewed. OM incidence, clinical features, and associated risk factors were analyzed.

Results

128 patients (23.2%) developed OM, peaking 7 days after stem cell infusion. Autologous transplantation reduced OM risk by 84.5%, while periodontitis and prior oral symptoms during chemotherapy increased risk by 111.4% and 140.7%, respectively. OM was associated with longer hospitalization and more febrile days. Symptoms were managed effectively with mouthwash, antibiotics, and ultraviolet therapy.

Conclusions

OM is common in HSCT patients and can be effectively controlled. Allogeneic transplantation, pre-existing oral conditions, and periodontitis increase OM risk, highlighting the importance of pre-transplant oral care.

Introduction

Oral mucositis (OM) is a common oral complication closely associated with hematopoietic stem cell transplantation (HSCT). It typically manifests as inflammatory or ulcerative lesions on the oral mucosal epithelium, ranging from mild erythema and atrophy to severe ulcerative mucosal destruction [1, 2]. OM has a high incidence rate and often leads to pain during eating, swallowing difficulties, and taste disturbances. These symptoms not only hinder oral intake but also result in dehydration, malnutrition, and decreased quality of life [3]. Moreover, OM can lead to secondary swelling and bleeding. Due to reduced gag reflex, it may increase the risk of airway obstruction and aspiration pneumonia, as well as the occurrence of opportunistic infections [1, 4].

Therefore, this study aims to analyze the incidence, risk factors, and clinical consequences of mucositis in patients undergoing HSCT during their transplant hospitalization. The objective is to elucidate the impact of mucositis on HSCT and to establish risk assessment criteria and preventive measures for mucositis.

Materials and methods

Patients diagnosed with hematological disorders who underwent pre-transplant oral examinations at the Department of Stomatology, Peking University People’s Hospital, from June 2019 to June 2022 were selected for this study. HSCT was performed before June 2023. Data collection was conducted from July 2023 to February 2024. The medical records of these patients, including pre-transplant oral examination records and transplant hospitalization records, were reviewed. General information (gender, age, type of hematological disorder), specialized oral examination and treatments before transplantation, number of pre-transplant chemotherapy sessions, medication details, presence of oral symptoms during chemotherapy and transplantation, duration of hospitalization, episodes of fever, occurrence of bacteremia, and the development of Acute Graft-Versus-Host Disease (Acute GVHD) [5] after transplantation were collected.

Patients with hematological disorders who had undergone HSCT, were aged 18 years or older in medical records, and had complete medical records were included in the analysis.

Exclusion criteria: (1) Patients without hematological disorders; (2) Patients who did not undergo HSCT until June 2023; (3) Patients with incomplete oral or hospitalization records.

The pre-transplant oral examination assessed the presence of oral lesions in patients, focusing primarily on endodontic diseases, periodontal diseases, mucosal disorders, and impacted third molars.

Definitions

Endodontic diseases [6] include dental caries, pulpal diseases, periapical diseases, and non-caries hard tissue diseases of dental pulp. The diagnosis of endodontic diseases involved the presence of impacted third molars combined with pulp diseases, residual roots and crowns, and apical lesions in affected teeth. Patients with teeth that had undergone complete filling or root canal treatment without symptoms and signs were not categorized under endodontic disease group.

Periodontal diseases [7] referred to inflammatory and destructive diseases involving the periodontal supporting tissues. Definition of gingivitis is full-mouth BOP(Bleeding on Probing) ≥ 10%, probing depth ≤ 3 mm, absence of attachment loss on adjacent surfaces or buccolateral/lingual sides, and no bone resorption.Periodontitis is characterized by the loss of periodontal tissue support, which is manifested by clinical attachment loss (CAL), radiographic alveolar bone loss, probing depth > 3 mm, and gingival bleeding.

Impacted Third Molars (ITM) [8] were defined as third molars that partially erupted or failed to erupt completely due to obstacles posed by adjacent bones or soft tissues. ITMs may be definitively judged as having lost eruptive potential when imaging, particularly CBCT, demonstrates (1) severely insufficient retromolar space, (2) horizontal, inverted, or pronounced mesioangular orientation, (3) near-complete root development with total bony impaction. Patients in the ITM group had one or more third molars. Patients in the non-ITM group did not have any third molars meeting the above definition, including those with only normally positioned third molars and those without third molars.

Clinical features of OM were recorded using toxicity grading of OM according to WHO [9].

Grade 0: Absence of other criteria;

Grade 1: Oral soreness and erythema;

Grade 2: Ulcers but able to eat solids;

Grade 3: Oral ulcers and able to take liquids only;

Grade 4: Oral alimentation impossible.

All hospitalized patients underwent oral assessment. Due to limitations in medical records, it is not possible to determine the specific grade of OM patients.

Human ethics declaration

As a retrospective study, we collected previous cases data and the process of analysis was anonymous. Patients’ private information was not exposed in the process of data analysis and paper writing. Informed consent was not available. In addition, our study has been approved by the Ethics Committee of Peking University People’s Hospital in accordance with the Declaration of Helsinki, and the approval number is IRB(2023PHE037-001).

Sample size calculation

Sample size calculation was performed using the formula:  

Allowable error d = 5%, α = 0.05, thus n ≈ (1.96² × 0.5 × 0.5) / (0.05²) ≈ 384.1.

The final cohort therefore exceeded the minimum target required for descriptive analysis.

Statistical analysis

Data analysis was performed using SPSS 25.0. For continuous variables, normality was assessed by histogram inspection and the Kolmogorov-Smirnov Lilliefors test. A P value > 0.05 was considered indicative of a normal distribution. Normally distributed data were compared with the t test, whereas non-normally distributed variables, including age, number of chemotherapy sessions, duration of transplant hospitalization and days of fever were analyzed with the Mann-Whitney U test. Chi-square tests were used to analyze gender, transplantation type, history of chemotherapy, presence of oral symptoms during chemotherapy, existing oral issues and bacteremia, and abnormal oral microbiota during transplantation, occurrence of Acute GVHD. Fisher’s exact test was employed for categories with sample sizes less than 40. A significance level of P < 0.05 was considered statistically significant.

For patient gender, transplantation type, periodontitis, occurrence of oral symptoms during chemotherapy, a multivariate logistic regression analysis was conducted to identify risk factors associated with the development of mucositis during transplant hospitalization.

Results

A total of 551 valid cases were included in the study, comprising 320 males and 231 females. Among these, a total of 128 patients were clearly diagnosed with mucositis, as documented in the medical records. General information about patients with and without mucositis is presented in Table 1.

Table 1.

Comparison of general conditions between OM and non OM patients

		OM(%)	Non OM(%)	p
Total		128(23.2)	423(76.7)
Male		75(58.59)	245(57.92)	0.892
Female		53(41.41)	178(42.08)
Age		40.70 ± 12.25	41.15 ± 12.99	0.753
Hematologic disease	AML	52(40.63)	142(33.57)
	ALL	37(28.91)	99(23.40)
	MM	2(1.56)	43(10.17)
	MDS	16(12.50)	64(15.13)
	Aplastic anemia	6(4.69)	19(4.49)
	Lymphoma	8(6.25)	31(7.33)
	Others	7(5.47)	25()5.91
Type of transplantation	Autologous	3(2.34)	59(13.95)	0.000
	Allogeneic	125(97.66)	364(86.05)
Type of regimens conditioning in allogeneic transplantation	Reduced-toxicity	24	86	0.306
	Standard-toxicity	101	278
Pre-transplantation chemotherapy	Total	117(91.4)	356(84.2)	0.039
	Experiencing oral symptoms during the chemotherapy	85/117(72.6)	187/356(52.5)	0.000
	Number of chemotherapy	3.93 ± 1.83	4.44 ± 2.48	0.246

From Table 1, significantly higher proportions of OM were noted in allograft patients compared to autologous transplant patients.Among the 551 patients, 473 individuals (85.84%) had undergone chemotherapy before transplantation, with 272 patients (57.51%) experiencing oral symptoms. Ulcers and mucositis were the most frequently occurring symptoms, accounting for approximately 63% of all patients with chemotherapy-related oral complications. Gingival swelling and pain, inflammation of wisdom teeth, toothache, gingival bleeding, salivary gland swelling, dry mouth, and tooth sensitivity occurred at lower frequencies. Patients who had undergone chemotherapy before transplantation and experienced oral symptoms showed a significantly higher incidence of mucositis in transplantation. The number of chemotherapy sessions did not have a significant impact on the incidence of mucositis (see Table 1).

Table 2 displays the influence of patients’ oral health status on OM. Patients with periodontitis had a significantly higher incidence of OM. There was no significant difference in the occurrence of mucositis between patients with treated periodontitis and those without periodontitis (p = 1). Mucositis occurred exclusively in patients with untreated periodontitis. Endodontic diseases, as well as impacted teeth, did not affect the incidence of OM.

Table 2.

Relationship between oral lesions and mucositis

Number of patients	OM	%	Non OM	%	p
Oral Lesions	117	91.41	375	88.65	0.377
Endodontic disease	52	40.63	184	43.50	0.565
Periodontal disease	98	76.56	265	62.65	0.004
Impact tooth	40	31.25	145	34.28	0.525
No history of oral treatment in the past year	93	72.66	342	80.85	0.046

The risk factors for developing mucositis are presented in Table 3. Autologous transplant patients had an 84.53% reduced risk of developing mucositis. Patients with periodontitis had a 111.42% increased risk of developing mucositis, and those with oral symptoms during chemotherapy had a 140.74% increased risk.

Table 3.

Risk factors analysis of oral mucosal inflammation during transplant hospitalization

	OR	95% Confidence Interval	P	B
Sex(male)	0.985	0.6322,1.5334	0.945	0.0155
Autologous	0.155	0.0465,0.5146	0.002	1.8662
Periocontal disease	2.114	1.2891,3.4677	0.003	0.7487
Experiencing oral symptoms during the chemotherapy	2.41	1.5093,3.8398	0.000	0.8785

Compared to patients without mucositis, mucositis patients experienced a significant increase in the duration of hospitalization during transplantation (P = 0.027), a significant increase in the number of febrile days during hospitalization (P = 0.033), and might have a significantly elevated risk of developing acute GVHD (P = 0.050). However, there were no significant differences observed in the occurrence of bacteremia or abnormal oral microbial flora among patients.

As shown in Fig. 1, nearly 50% of OM cases occurred between 5 and 8 days after stem cell infusion, with a peak incidence observed 7 days after cell infusion. At the onset of OM, the mean white blood cell count was 0.512 ± 1.374*10^9/L, and the mean neutrophil count was 0.329 ± 1.142*10^9/L. Among these, 108 patients (84.62%) had neutrophil counts below 0.1*10^9/L, while only 8 patients (6.25%) had neutrophil counts above 1*10^9/L, with data collected within 24 h of the clinical diagnosis of OM.

Fig. 1.

Time of onset of mucositis

Mucositis in the oral cavity predominantly presented as multiple lesions (66 individuals), with a higher incidence in the pharynx (38 individuals), cheeks (32 individuals), tongue (29 individuals), and a lower occurrence on the lips (5 individuals) and gums (14 individuals). The majority of mucositis patients (93 individuals) alleviated symptoms through intensified mouthwash, supplemented with antibiotic therapy in 59 cases, ultraviolet irradiation in 47 cases, and some patients used other topical applications (32 individuals) for local treatment.

Discussion

Oral mucositis (OM) is a major non-hematologic complication of HSCT (HSCT) [10]. In the general overview of 551 HSCT patients, we found no significant correlation between the incidence of OM and age or gender in adult patients. However, some previous studies reported a higher incidence of OM in younger patients [11, 12]; the difference in our findings might be attributed to the fact that our study population consisted of adults aged 18 and above, excluding pediatric patients included in those studies. Prior literature lacked relevant statistical analyses regarding the impact of transplantation methods on the incidence of mucositis. We found a significant difference in the incidence of mucositis based on different transplantation methods. Allogeneic transplant patients had a significantly higher proportion of OM. This could be due to the fact that autologous transplant preconditioning regimens have lower drug toxicity and intensity, reducing mucosal irritation. Additionally, autologous transplant patients do not have the risk of developing GVHD, eliminating the need for routine prophylactic use of methotrexate, which, in turn, avoids adverse reactions associated with methotrexate that can promote oral ulceration [13].

Our data revealed no significant association between the type of hematologic disorder and the incidence of OM. However, a study indicated that patients with malignant hematologic disorders (various types of leukemia, lymphoma, etc.) had a lower risk of mucositis compared to patients with immunodeficiency syndromes or hereditary hematologic diseases [12]. It is important to note that this study was based on data from 170 pediatric patients, which differs substantially from our adult patient population. The potential impact of different hematologic disorders on mucositis in pediatric and adult patients warrants further investigation in our future studies.

There is ongoing controversy regarding the impact of different transplant preconditioning regimens on mucositis. One study suggested that multidrug chemotherapy regimens, especially those involving multiple alkylating agents, lead to more severe OM compared to single-agent regimens [4]. Other literature indicates that myeloablative conditioning regimens and drugs used for GVHD prevention, such as methotrexate, are important risk factors for OM [13]. However, a study involving 640 patients found that reduced-intensity conditioning (RIC) resulted in a similar high incidence of OM compared to conventional regimens, suggesting that the intensity of preconditioning does not affect the incidence of OM [14]. In our hospital, most allogeneic transplant patients (including various types of leukemia, lymphoma, MDS, etc.) undergo myeloablative conditioning. Patients at higher risk and the majority of relapsed or refractory patients receive reduced-toxicity conditioning regimens. However, we have not observed a significant association between standard or reduced-toxicity conditioning regimens and the incidence of mucositis. Non-myeloablative regimens are mainly used for autologous transplant patients with diseases such as multiple myeloma, as mentioned earlier, leading to a significantly reduced incidence of mucositis in these patients.

Previous studies have rarely mentioned the correlation between pre-transplant chemotherapy and OM during transplantation. In our research, we collected relevant information about patients’ pre-transplant chemotherapy and oral symptoms, which were often overlooked in previous studies. Among patients who developed OM during transplantation, the proportion of those who had undergone chemotherapy before transplantation was significantly higher than in the non-OM group. Moreover, the percentage of patients with oral symptoms related to previous chemotherapy was also notably higher. This indicates that a history of oral symptoms during chemotherapy was associated with a higher incidence of OM during HSCT. Since chemotherapy drugs have similar effects to conditioning agents, patients who experienced OM during pre-transplant chemotherapy are at a significantly higher risk of developing mucositis during transplantation. This suggests a heightened individual susceptibility to the mucosal toxicity of chemotherapeutic agents. Patients who exhibited mucosal vulnerability during induction or consolidation chemotherapy may harbor biological or genetic predispositions that lower their threshold for conditioning-induced mucosal injury during HSCT. It is crucial to focus on these patients during transplantation and implement preventive measures and enhanced oral care promptly.

The existing oral health status of patients and whether prior dental treatments impact the occurrence of OM are also worth discussing. One study found no significant correlation between impacted or partially impacted teeth and OM [15], aligning with our conclusion. Moreover, we confirmed for the first time that dental conditions such as endodontic diseases are not significantly associated with the incidence of mucositis. However, there is a debate regarding the impact of patients’ periodontal health status on OM. Specifically, questions arise about whether pathological conditions like periodontitis and gingivitis increase the incidence of OM and whether pre-transplant periodontal treatments are necessary.Some studies argue that periodontal health has no statistically significant influence on the occurrence of OM. The prevalence and severity of oral mucositis after HSCT are not affected by pre-transplant oral hygiene and the presence of periodontal diseases [3]. In contrast, other research suggests that patients with periodontal diseases have a higher incidence of OM. After periodontal treatment, improvements in periodontal health lead to a significant reduction in the incidence and severity of OM [1, 16]. Our findings support the latter conclusion. We observed a significant increase in the likelihood of mucositis in patients with untreated periodontitis compared to those without periodontitis. It is noteworthy that none of the 4 patients who received professional periodontal treatment prior to HSCT developed OM. The small sample size of the treated group may have resulted in a false negative (Type II error). Future prospective studies with larger cohorts are required to validate the protective effect of pre-transplant periodontal intervention.

In this study, the results showed that the incidence of mucositis in the 551 included patients was 23.2%, which is relatively lower than other studies. Previous literature has indicated mucositis incidence rates ranging from 50% to 100% [3, 11, 13, 14, 17–19]. This discrepancy might be due to potential missing or overlooked records and descriptions related to mucositis in the medical records. Mild cases of mucositis might not have been documented .Besides, age and the type of transplantation mode may also affect the incidence of mucositis. Some previous studies reported a higher incidence of OM in younger patients [11, 12], our study population consisted of adults aged 18 and above, excluding pediatric patients included in those studies. Second, allogeneic transplant patients had a significantly higher proportion of OM than autologous ones. Among the 551 patients in our study, 62 underwent autologous transplantation.In conclusion, both the inclusion of older patients and autologous transplantation patients will reduce the incidence of mucositis. Additionally, accurate statistics on the duration and grading diagnosis of mucositis were challenging due to incomplete oral records. Hence, prospective studies are necessary to improve further analyses in the future.

In our study, the peak onset of OM occurred between 5 and 10 days after stem cell infusion. This finding is consistent with other literature reports [3, 4]. During the onset of OM, patients had extremely low white blood cell and neutrophil counts. Nearly 85% of OM patients had neutrophil counts below 0.1*10^9/L, indicating they were in a state of extreme immune suppression. No significant differences in mucositis incidence were found among different mucosal sites. There is limited literature available on the specific sites of mucositis onset. Some studies have shown a correlation between longer hospital stays and mucositis [5, 13, 17], and the presence of OM significantly increases the occurrence of fever and leads to increased use of analgesics and parenteral nutrition [13], which is similar to our study. This adds to the patients’ treatment burden.

Some researchers have proposed that Streptococcus is the most common potential oral source of bacteremia [19], and ulcerative OM is a common complication of allogeneic transplantation, providing a potential entry point for Streptococcus [20]. In this study, only 23 cases of mucositis patients had abnormal oral flora, out of which 4 cases were associated with bacteremia. The incidence of bacteremia during transplantation did not show a statistically significant difference between patients with and without abnormal oral flora. Similarly, there was no significant difference in the occurrence of bacteremia between patients with and without mucositis, indicating that OM does not increase the probability of bacteremia. This finding aligns with some previous literature conclusions, the severity and duration of OM are not related to bacteremia [17].

Analyzing the relationship between mucositis and acute GVHD, our study showed a significantly higher proportion of acute GVHD in patients with mucositis compared to those without mucositis. This suggests that OM was associated with a higher incidence of acute GVHD, consistent with previous studies [12].

Regarding the treatment of mucositis, it is common practice to use calcium folinate mouthwash after stem cell infusion. This is done to counteract the toxic side effects of methotrexate on the oral and digestive mucosa. Additional mouthwashes, including chlorhexidine, sodium bicarbonate, lidocaine, nystatin, gentamicin, recombinant human interleukin-11 (rhIL-11), and cetylpyridinium chloride, are used based on the presence of fever and blood test results in patients with mucositis. Gentamicin mouthwash, which contains recombinant human interleukin-11, has been shown in research to significantly shorten the healing time of oral mucositis [21].Antibiotics such as cefoperazone sodium and sulbactam sodium, imipenem and cilastatin sodium, amikacin, tigecycline, cefoperazone sodium and sulbactam sodium, meropenem, tazobactam, vancomycin hydrochloride, meropenemand piperacillin sodium and tazobactam sodium are used based on the presence of fever and blood tests. Topical treatments include iodine glycerin and recombinant human epidermal growth factor ( rhEGF).

After these treatments, all cases of oral mucositis improved. No cases were found where OM interrupted the transplantation process or caused severe life-threatening infections. Therefore, the incidence and severity of OM do not affect overall survival and transplant success rates, aligning with previous research findings [22].

Current research in OM treatment primarily focuses on using cytokines and growth factors to reduce the toxic effects on the oral mucosa and promote faster re-epithelialization [4, 23, 24].

Limitations

This was a retrospective study, and reliance on medical records may have introduced incomplete information and information bias. First, the retrospective single-center design may limit the generalizability of the findings. Although the sample size was adequate for a short-term retrospective analysis, the observational design and limited study period restricted our ability to control for multiple confounders. Second, the absence of systematic bedside dental examinations during HSCT probably led to underestimation of the true incidence of mucositis, and reliable WHO grade-specific analyses could not be performed. Besides, only a small number of patients received professional periodontal treatment before HSCT, limiting the statistical power to evaluate the protective effect of periodontal intervention. And there is potential residual confounding.

Future prospective studies should adopt structured recording protocols based on the WHO mucositis grading system. Analyses should further clarify the effects of age, medications, treatment regimens, and pre-existing oral disease on complications such as mucositis during transplantation. Long-term prospective follow-up is also needed.

Conclusion

Oral mucositis (OM) is a common occurrence in patients undergoing HSCT. Most cases manifest as multiple lesions within the oral cavity, with the peak incidence observed between 5 and 10 days after stem cell infusion. At this stage, patients often exhibit severely reduced neutrophil and leukocyte counts. OM symptoms can be effectively controlled through treatment, without adversely affecting overall survival and transplant success rates. However, patients with mucositis experience prolonged hospitalization and an increased number of days with fever in the transplant unit. OM was associated with a higher incidence of acute GVHD.

Allogeneic transplant recipients, as well as those experiencing oral symptoms during chemotherapy or individuals diagnosed with periodontitis, face an elevated risk of developing OM within the transplant unit. Therefore, it is advisable to focus on pre-transplant interventions for patients scheduled for allogeneic transplantation. In cases where oral symptoms arise during chemotherapy, pre-transplantion periodontal treatment is recommended to alleviate gingival inflammation, thus reducing the likelihood of OM occurrence.

Authors’ contributions

Xuan Gao, Conceptualization; Investigation; Data curation; Formal analysis; Methodology; Writing – original draft; Writing – review & editing. Shengnan Zhao, Conceptualization; Investigation; Data curation; Validation; Writing – original draft; Writing – review & editing . Chengzhi Gao, Conceptualization; Project administration; Supervision. Peng Wu, Conceptualization; Investigation. Lijuan Wang, Methodology; Data curation. Zixin Li, Writing – review & editing, Data curation. Panhui Chang: Supervision; Data curation; Formal analysis; Writing – review & editing.

Funding

This research received no external funding.

Data availability

The datasets used during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Approved by the Ethics Committee of Peking University People's Hospital in accordance with the Declaration of Helsinki (IRB 2023PHE037-001).

Not available (As a retrospective study).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.