This was a cross-sectional survey study aimed at collecting patients' recommendations on the optimal cold compress duration for pain relief and quality of life improvement after mandibular impacted third molar extraction.

The authors conducted a survey on 60 patients who had undergone mandibular impacted third molar extraction at the Department of Oral and Maxillofacial Surgery.

A questionnaire was first designed, including patients' age, sex, action of postoperative cold compress, cold compress material, cold compress method and whether frostbite has occurred. The questionnaires were filled out based on the patients' conditions.

This was a pilot study designed to simulate real-world clinical scenarios and prevent cryotherapy-related adverse events. It also aimed to refine protocols for subsequent randomized controlled trials (RCTs).

A total of 30 cases were involved in this pilot study. These patients were selected from those who had undergone mandibular impacted third molar extraction at the Department of Oral and Maxillofacial Surgery.

Similar to the later RCTs, cold compress therapy was applied, and skin temperatures and safety were closely monitored during the process.

The endpoints included the incidence of frostbite and other adverse events related to cold compress, as well as the feasibility of the cold compress protocols in a real-world setting. The findings were used to optimize the protocols for the subsequent RCTs.

This was a prospective randomized controlled clinical trial comparing the effects of intermittent and continuous cold compress on patients after mandibular impacted third molar extraction.

A power analysis using the G*Power 3.1.9.7 software was conducted. Under the set α value of 0.05 and a 90% power analysis condition, each group (intervention and control) required 38 patients as the sample size.

The stratified randomization method was adopted, with the number of teeth extracted (one or two) as the stratification variable. After determining the stratification variables, a random allocation sequence for different stratifications (namely, extracting one tooth and extracting two teeth) was generated using a computer. To ensure the confidentiality of the allocation, a series of security measures were taken for the generated random allocation sequence. Each random allocation sequence within each stratum was given a consecutive number, and then the numbered sequences were placed separately into sealed and opaque envelopes. These envelopes were strictly kept before the inclusion of patients who met the inclusion criteria to prevent information leakage. During the patient recruitment process, when a patient met the inclusion criteria, their corresponding stratification group was determined based on the actual number of teeth extracted. Then, the opaque envelope corresponding to that stratification group was opened, and the patient was assigned to the appropriate group according to the random allocation sequence indicated in the envelope. Eventually, after all 76 patients who met the inclusion criteria completed the allocation, they were randomly and evenly divided into the control group and the intervention group, with 38 patients in each group.

Inclusion Criteria: Patients meeting the indications for third molar extraction with no absolute contraindications; Diagnosis of mandibular impacted third molar confirmed by clinical oral examination and panoramic/x-ray imaging, requiring osteotomy, flap reflection, and removal of dental/bony obstructions during extraction; Surgical duration <2 h, with no intraoperative root fracture or adjacent tooth injury.

Exclusion Criteria: Allergy to local anesthetics (e.g., lidocaine) or use of postoperative analgesic interventions; Positive cold sensitivity test; History of diabetes mellitus or coagulation disorders; Peripheral circulatory insufficiency or sensory neuropathy affecting pain perception; Midway withdraw from the study.

All patients were advised to perform mandibular impacted third molar extraction after having breakfast in the morning. After the visit, panoramic x-rays were taken for all patients. Before the operation, the difficulty of the surgery and the risks during and after the procedure were analyzed. Each patient signed the informed consent form for tooth extraction at the Department of Oral and Maxillofacial Surgery, The Second Hospital of Hebei Medical University. All the surgeries were performed by the first author of this study alone. Routine disinfection, dressing, anesthesia, incision and flap lifting, and extraction of impacted teeth were carried out. All patients had panoramic x-rays taken after the surgery to confirm that there were no remaining tooth fragments or bone fragments. After the complete extraction of the teeth, the patients were instructed to take antibiotics including cefadroxil tablets (0.5 g/time, twice a day; CSPC Ouyi Pharmaceutical Co., Ltd) and ornidazole dispersible tablets (0.25 g/time, twice a day; HENAN TOPFOND PHARMACEUTICAL CO., LTD) orally for 3 days.

During each operation, 1–2 teeth were extracted and a unilateral approach was adopted. After the teeth on one side were extracted, factors such as the difficulty of the extraction, the patient's recovery condition, and overall health status were taken into consideration to determine the timing of extracting the teeth on the other side. It was usually recommended to wait 2 to 4 weeks, but the specific time was determined by the doctor based on the individual situation.

Control group: Intermittent cold compress protocol: 30 min cold compress sessions alternating with 1 h intervals, repeated for 6 h.

Intervention group: Continuous cold compress applied uninterrupted for 6 h. Before implementing continuous cold compress, we conducted a comprehensive assessment of all the patients in this group, inquiring and recording their past medical history in detail, with particular attention to whether there were any dangerous conditions related to prolonged cold compress, such as local circulatory disorders (such as Raynaud's disease), abnormal skin sensation disorders, history of cold allergy, local skin damage or infection. After the assessment, it was found that all the patients included in the continuous cold compress group in this study had no such medical history that might aggravate the dangerous conditions caused by continuous cold compress. During the cold compress process, a dedicated person was arranged to regularly observe the local skin condition of the patients. If any abnormal reactions such as pale skin, cyanosis, numbness, or increased pain occurred, the cold compress was immediately stopped and corresponding treatment measures were taken.

To ensure the stability and operability of the cold compress effect, in this study, medical ice packs were uniformly used as the cold compress material.

The endpoints were evaluated immediately, and at 1.5, 3, 4.5, and 6 h postoperatively. The main outcome measures included pain intensity, facial swelling, maximum mouth opening, and wound hemorrhage.

This was a prospective randomized controlled clinical trial aimed at optimizing the duration of cold compress application after mandibular impacted third molar extraction.

A power analysis using the G*Power 3.1.9.7 software was conducted. Under the set α value of 0.05 and a 90% power analysis condition, each group required 20 patients as the sample size.

A group randomization design was adopted. The random allocation sequence was generated by a computer. The confidentiality of the allocation was ensured through measures such as consecutive numbering, sealing, and opaque envelopes. After meeting the inclusion criteria, 80 patients were randomly divided equally into Group A, Group B, Group C and control group.

The inclusion and exclusion criteria were the same as those in Sub-study 3.

The surgical methods were identical to those described in Sub-study 3.

All intervention subgroups initiated cold compress immediately postoperatively. Patients returned for follow-up every other day, with scheduled visits timed to match the original extraction appointment to minimize circadian variability.

The endpoints were assessed on postoperative days D1-D3 for pain intensity, facial swelling, maximum mouth opening, and wound hemorrhage.

Visual Analog Scale (VAS)/Numerical Rating Scale (NRS) (18) have good reliability and validity. The Cronbach reliability coefficient of VAS and NRS was 0.845 and 0.830, respectively (19, 20): 0: No pain; 1–3: Mild pain; 4–6: Moderate pain; 7–10: Severe pain.

Facial swelling percentage was calculated using the formula: Facial swelling (%) = Post-compression measurement—Baseline measurement/Baseline measurement × 100% (21). Baseline measurement: Average of two facial distances (cm): Mandibular angle to lateral canthus and Oral commissure to earlobe. Swelling Grading: Grade 0: ≤3%; Grade I: 3%–6%;Grade II: 6%–12%;Grade III: >12%.

Normal: 3.5–4.5 cm; Grade I (Mild trismus): 2–2.5 cm; Grade II (Moderate trismus): 1–2 cm; Grade III (Severe trismus): <1 cm; Grade IV (Complete trismus): Lockjaw (22).

Grade 0: No bleeding; Grade 1: Pink-tinged saliva; Grade 2: Bright red saliva; Grade 3: Frank blood or clots (23).

The CONSORT diagram was shown in Figure 3. There were no significant differences in gender, age, duration of surgery and number of mandibular third molars extracted between the two groups (P > 0.05, Table 2), indicating the general information of the two groups were comparable.

As shown in Table 3, there were no significant differences in VAS scores between the two groups at pre-intervention and 1.5 h, 3 h and 4.5 h post-intervention (P > 0.05). However, the VAS scores of the intervention group were lower than those of the control group 6 h post-intervention (P < 0.001).

As shown in Table 4, there was no significant difference in the frequency of facial swelling between the two groups at pre-intervention, 1.5 h, 3 h, 4.5 h, and 6 h post-intervention (P > 0.05).

There was one patient with limited mouth opening in the intervention group 6 h post-intervention.

As shown in Table 5, there was no significant difference in wound bleeding between the two groups 1.5 h, 3 h and 4.5 h post-intervention (P > 0.05). However, the frequency of patients with grade I wound bleeding in the intervention group was lower than that in the control group 6 h post-intervention (P = 0.047).

The CONSORT diagram was shown in Figure 4. There were no significant differences in gender, age, duration of surgery and number of mandibular third molars extracted among the four groups (P > 0.05, Table 6), indicating the general information of the four groups were comparable.

As shown in Tables 7–9, the VAS scores of the experimental subgroups were lower than those of the control group at postoperative Day 1 (P = 0.01). However, there were no significant differences in the VAS scores between the experimental subgroups and control group at postoperative Day 2 and Day 3 (P > 0.05).

There was no significant difference in the frequency of facial swelling between the experimental subgroups and control group at postoperative Day 1 and Day 2. However, compared with the control group, the frequency of facial swelling in the experimental subgroups was lower at postoperative Day 3 (P = 0.001).

There was no significant difference in the frequency of limited mouth opening between the experimental subgroups and control group at postoperative Day 1 and Day 3. However, compared with the control group, the frequency of limited mouth opening in the experimental subgroups was lower at postoperative Day 2 (P = 0.002).

Compared with the control group, the frequency of patients with grade I wound bleeding in the experimental subgroups was lower at postoperative Day 1 (P = 0.003). However, there was no significant difference in the frequency of patients with grade I wound bleeding between the experimental subgroups and control group at postoperative Day 2 and Day 3 (P > 0.05).

This study demonstrated that continuous cold compress application within the first 6 postoperative hours significantly outperformed intermittent protocols in analgesia, edema reduction, trismus alleviation, and hemorrhage control. The underlying mechanisms are elucidated through three key pathways:

Based on the findings, a staged cold compress protocol is proposed: Phase I (0–6 h postoperatively): Continuous cold compress application, prioritizing acute hemorrhage control and pain mitigation; Phase II (D1–D3): Daily 6 h continuous cold compress sessions to suppress inflammatory edema and improve trismus resolution.

In recent years, several new postoperative management techniques have emerged in tooth extraction surgery, including platelet-rich fibrin (PRF), ozone therapy, low-level laser therapy (LLLT) and kinesiology taping (33–36). PRF is known for its regenerative and anti-inflammatory properties, and thus increasingly being used to reduce postoperative pain, swelling, and symptoms of alveolar osteitis (37). Compared with PRF, cold compresses do not require invasive procedures, are inexpensive, and can be applied immediately. However, PRF achieves long-term effects through biological repair, while cold compresses only provide short-term symptom relief. The two can complement each other (for example, early cold compress after surgery to control acute symptoms, and later filling with PRF to promote healing). Ozone therapy has shown promise in minimizing postoperative discomfort and promoting healing due to its antimicrobial and oxygenating effects (38). Compared with ozone therapy, cold compresses have no odor and no injection risks, and can be operated independently; ozone therapy, on the other hand, requires professional training from medical staff and has a low equipment penetration rate. However, ozone therapy achieves comprehensive effects by improving the micro-environment, while cold compresses only address the symptoms. The two can be used together (for example, ozone rinse followed by cold compress to reduce exudation). LLLT inhibits the expression of COX-2 through photobiological regulation, reduces the synthesis of PGE2, and simultaneously promotes mitochondrial ATP production, thereby alleviating muscle spasms and pain (39). Compared with LLLT, cold compresses are extremely cost-effective and can be applied immediately; LLLT requires specialized equipment and its effectiveness is influenced by the operator's experience. However, LLLT achieves deep tissue repair through biological stimulation, while cold compresses only act on the surface tissues. The two can work together (for example, after cold compresses control acute symptoms, LLLT promotes functional recovery). Kinesiology taping is an innovative non-invasive therapy that has been proven to alleviate swelling after oral surgery and facilitate the drainage of lymph fluid (40). Compared with kinesiology taping, cold compresses do not require special training and can cover large areas; kinesiology taping, on the other hand, need to be custom-cut and may cause skin allergies. However, kinesiology taping improves lymphatic circulation through mechanical action, while cold compress works through thermodynamic effects. The two can be combined (for example, applying the kinesiology taping after cold compressing to prolong the time for reducing swelling). By comparison, it can be seen that the “6 h continuous cold compress protocol” proposed in this study has irreplaceable advantages in terms of accessibility, cost and short-term efficacy, and is particularly suitable for promotion in grassroots medical institutions. At the same time, the research results provide a theoretical basis for the multimodal postoperative management of “cold compress + emerging technologies”, promoting the clinical transition from “single treatment” to “integrative medicine” mode. In the future, a head-to-head randomized controlled trial needs to be conducted to clarify the optimal combination sequence and timing of different technologies, in order to further optimize the nursing path.