Can cryoprevention of oral mucositis be obtained at a higher temperature?

Cryotherapy (CT) has for long been used in various medical disciplines to manage a variety of life-threatening conditions, e.g. major stroke, severe traumatic brain injury and traumatic cardiac arrest [1]. CT has over the past decades further been applied in haematology/oncology to reduce the duration and severity of chemotherapy-induced oral mucositis (OM), a common debilitating sequel to haematopoietic stem cell transplantation (HSCT) [2‐5].

OM refers to the painful, erythematous and ulcerative lesions in the oral mucosa which affects 40% of patients who receive standard dose of chemotherapy, and up to 80% of patients who are conditioned with high-dose chemotherapy in preparation for HSCT [6, 7]. Once established in the oral mucosa, OM can give rise to several distressing symptoms. These include oral pain, dysphagia and the need for parenteral nutrition, all of which may adversely affect the patient’s well-being. As a consequence, this may lead to increased need for analgesics, extended hospital visits [8‐10] and entail increased healthcare costs [11, 12]. OM can further be a portal of entry for systemic infections which can lead to sepsis and death [13, 14].

Current recommendations by the Multinational Association of Supportive Care in Cancer/International Society of Oral Oncology (MASCC/ISOO) for prevention of chemotherapy-induced OM include recombinant human keratinocyte growth factor-1 (KGF-1, palifermin), photobiomodulation and CT [4]. The latter comprises the largest proportion of studies and continues to be the best alternative of the three strategies [4]. Several studies, including randomised controlled trials, have consistently favoured the use of CT for prevention of chemotherapy-induced OM [5, 15, 16]. However, despite these positive observations, the use of ice cooling as a preventive method in clinical practice is limited. The reason for this may be that ice has a detrimental effect on the comfort level of the patients, e.g. by causing cold sensations, chills and shooting pain in the teeth, and thereby leading to poorer adherence to IC [17, 18]. In addition, the use of IC requires water of good quality, to avoid contamination by microorganisms, which may have an unfavourable effect for an already immunosuppressed patient. Therefore, there have been demands in the literature to improve the use of cooling devices for cryoprevention (CP) [18]. In response to this, an intraoral cooling device (ICD; Fig. 1a), which can be set to operate at different temperatures, has been developed. In a previous study [17], we observed that the ICD was better tolerated than IC and that the two cooling methods were equally effective with regard to intraoral temperature reduction. It is therefore of interest to elucidate if higher temperatures than those previously obtained by IC could enhance tolerability without compromising the intraoral temperature necessary to prevent chemotherapy-induced OM.

All subjects enrolled in this study were healthy and reported no medical conditions. Table 1 provides a summary of the subject characteristics, BMI (kg/m²), basic haemodynamics (heart rate, bmp; systolic and diastolic blood pressure, mmHg; mean arterial pressure, mmHg) and procedural times in the present study. In total, 40 cooling sessions were completed, and all participants endured the 60 min of cooling with the ICD set to both temperatures. Furthermore, all questionnaires were completed and a total of 960 intraoral thermographic images were captured during this study.

At baseline, prior to the cooling sessions with the ICD set to either 8 °C or 15 °C, mean intraoral temperatures were 35.2 °C and 35.0 °C, respectively (Fig. 2). This difference of 0.2 °C was not statistically significant.

Following 60 min of cooling, with the ICD set to either 8 °C or 15 °C, a statistically significant difference between the two observations was demonstrated. In the cooling sessions using 8 °C, a mean temperature reduction of 7.69 °C was observed, whereas the corresponding temperature reduction at the same follow-up period when using 15 °C was 5.21 °C (Table 2). Thus, the evaluation of the primary endpoint showed a difference in mean temperature reduction of 2.48 °C between the cooling sessions (p < 0.001; 95% CI 1.30 to 3.40) (Fig. 2).

The statistical analysis for the secondary endpoint, i.e. the difference in mean temperature reduction (°C) after 30 min of cooling between the ICD set to 8 °C or 15 °C, also revealed a statistically significant difference between the cooling sessions (Fig. 2). Mean temperature reduction in the cooling sessions using 8 °C was 6.89 °C, whereas the same figure for the cooling session using 15 °C was 5.02 °C (Table 2). Thus, there was a mean difference of 1.87 °C (p < 0.001; 95% CI 0.73 to 2.60) (Fig. 2).

Following 30 and 60 min of cooling with the ICD set to 8 °C showed a mean temperature reduction of 6.89 °C and 7.69 °C, respectively (Table 2). The difference in mean temperature reduction of 0.8 °C was not statistically significant (Table 3). The same analysis, i.e. after 30 and 60 min for 15 °C, was 5.02 °C and 5.21 °C, respectively (Table 2). This difference of 0.19 °C did not reach statistical significance (Table 3).

The questionnaire was completed by all subjects (n = 20), and the responses to “Which of the two cooling sessions did you tolerate better?”, were as follows: one (n = 1) individual preferred cooling with the ICD set to 8 °C, four (n = 4) had no preference (shared in running text) and rated the two temperatures as equally tolerable and 15 participants preferred cooling with the ICD set to 15 °C. Thus, 15 of the 20 subjects (75%) preferred cooling with the ICD set to the higher cooling temperature. This difference was statistically significant (p < 0.001; 95% CI 50.9 to 91.3).

Concomitant registration for subject-reported adverse events is presented in Table 4. Rubbing discomfort was the most commonly reported adverse event in both cooling sessions, for 8 °C (n = 12) and for 15 °C (n = 12). In addition, in the cooling sessions using the ICD set to 8 °C, the following were reported: cold (n = 8), difficulties swallowing (n = 8), and poor fit (n = 7). The corresponding figures for the cooling sessions using the ICD set to 15 °C was cold (n = 2), difficulties swallowing (n = 9) and poor fit (n = 8).

The intraoral mucosa is a highly vascularised and innervated structure, which serves to protect the underlying tissues from mechanical stress, microorganisms and harmful toxins [20]. Even minor disruptions of these functions can seriously jeopardise the integrity of the oral mucosa and pose an increased risk to immunocompromised patients. OM constitutes such a risk which may lead to a potentially fatal condition. Fortunately, the development of chemotherapy-induced OM can be effectively inhibited by CP. In addition to reducing severity and duration of OM, CP has also been observed to decrease the number of days with total parenteral nutrition [16, 21], need for analgesics [15, 16, 22], length of hospitalisation [21] and pain related to OM [16]. Regrettably, IC has not been adopted as expected. This is surprising as the two leading international organisations—MASCC/ISOO—have for long recommended CP for management of OM [4]. Thus, numerous patients are at risk of suffering from the adverse effects of OM. One possible explanation for this neglection is the obvious infectious risk making IC from contaminated tap water and thereby threaten the medical rehabilitation of an already immunosuppressed patient [23‐25]. This was the rationale for the development of a novel ICD [17].

The primary endpoint of this study measured the difference in mean temperature reduction after 60 min of cooling between the ICD set to 8 °C or 15 °C. Our results demonstrated a statistically significant difference at the follow-up between the two temperatures tested. However, although a statistically significant difference was seen, the clinical relevance remains questionable. Most likely, the additional temperature reduction of approximately 2 °C will not provide superior benefits in CP of OM.

As part of the secondary endpoint, the mean temperature reduction between the ICD set to 8 °C or 15 °C was also evaluated after 30 min of cooling and the obtained data for the respective temperatures was compared with the corresponding data following 60 min of cooling. This was carried out in order to investigate during what time interval, 0–30 min or 31–60 min, the most significant temperature reduction occurs, and if longer cooling times provide additional effects of relevance with regard to temperature reduction. Similar to the results for the primary endpoint, there was a statistically significant difference after 30 min of cooling between the ICD set to 8 °C or 15 °C. However, the difference was of a smaller magnitude and therefore probably of even less clinical significance.

As expected, the greatest effect in temperature reduction, regardless of the two temperatures tested, occurred within the first-time interval. The result indicates that 30 min of cooling with the ICD is sufficient to achieve a steady-state temperature in the intraoral mucosa. However, it is not evident from our study when during the initial 30 min the steady-state temperature is achieved. To establish this time-point is a topic of interest to be investigated in future studies. There are two main reasons for such a study to be conducted. Firstly, if shorter cooling sessions are sufficient to achieve the steady-state temperature, tolerability is most likely enhanced which would be advantageous. Secondly, present clinical studies differ markedly with respect to when to start cooling prior to the chemotherapy.

Surprisingly, in the majority of clinical studies and in a comprehensive literature review by the Cochrane collaboration, no or few adverse events have been reported for CT using IC [13, 16, 26]. However, these results are not in line with our previous study [17], where several adverse events, e.g. cold, headache, and teeth sensations, were reported in the sessions using IC. This study further showed that the novel cooling method which employs higher cooling temperatures (8 °C) than IC was significantly better tolerated and was preferred over IC by 80% of the participants. Given this, it was of interest as part of the secondary endpoints to assess the tolerability profile of higher cooling temperatures than those previously reported as well to assess adverse events for the ICD. In the present study, it was shown that higher cooling temperature was better tolerated by the majority of the subjects and that self-reported adverse events were more frequently reported in the sessions with the ICD set to 8 °C. These were mainly attributed to temperature-related factors, which is not surprising as lower temperatures are accompanied by increased discomfort [17]. However, in both sessions, regardless of cooling temperature used, adverse events related to the design of the ICD were frequently reported. Thus, it is essential to modify the design of the ICD prior to implementation in clinical settings.

However, it is worth mentioning that despite the complaints, the majority of the subjects reported that cooling did not cause problems to such extent that they could not undertake other activities during the time of cooling. This suggests that the overall perception of the ICD was acceptable. Given these results and the fact that there is room to further improve the design, the ICD can be considered for CP of OM in future clinical trials. This is particularly true in areas where the risk of contamination from water of poor quality is substantial [23‐25].

This study has some limitations that should be acknowledged. Firstly, a common weakness of crossover trials is the carryover effect. This is, however, unlikely to have affected the results of this study as a washout period of at least 24 h between the cooling sessions was imposed. Secondly, the thermographic camera is not specifically designed by the manufacturer for the purpose of this study. However, given that these potential limiting factors were similar in both intervention groups, it can be assumed that it has influenced the outcome to similar extent. It is noteworthy that the subjects in this study were dental students with no general health conditions or compromised oral health status. This is likely not the scenario in clinical settings with cancer patients, which may complicate the use of the ICD.

In conclusion, this study demonstrates that intraoral cooling using the ICD with a temperature of 15 °C is better tolerated than 8 °C but displays inferior capacity in temperature reduction of the oral mucosa. However, to elucidate whether this discrepancy of approximately 2 °C is of clinical importance, the optimal temperature for prevention of OM needs to be identified.