Sclerotherapy of ranulas with OK-432 – a prospective, randomized, double-blinded placebo-controlled study

Abstract

Background

Ranula is a rare benign cystic lesion in the floor of the mouth, which can herniate through the mylohyoid muscle and become a plunging ranula. Treatment for ranulas is currently surgical excision of the sublingual gland. Sclerotherapy with OK-432 is a well-established treatment of lymphatic malformations, but not yet thoroughly evaluated on ranulas. Objectives: To evaluate sclerotherapy of ranulas with OK-432 in a randomized double-blinded trial.

Materials and methods

20 patients with plunging or intraoral ranula were randomized to two double-blinded injections with OK-432 or saline. Effect on the ranula and evaluation of symptoms and QOL were investigated.

Results

Treatment response differed significantly between OK-432 and placebo, p = .041(student’s T-test). All patients with intraoral ranulas had a complete response, but only 1/4 of the patients with plunging ranula. The inflammatory reaction after injection with OK-432 caused a mild to moderate impact on QOL. No serious complications were observed.

Conclusion

This study suggests that sclerotherapy with OK-432 in ranula is a very effective treatment for intraoral ranulas, but possibly less useful in plunging ranulas.

Significance

This is a limited study, but we believe that sclerotherapy with OK-432 should be recommended as primary treatment at least for intraoral ranulas.

Chinese abstract

背景：舌下囊肿是一种口腔底部的罕见良性囊性病变, 可穿过髓样肌突起, 并形成一个下陷囊肿。目前, 对舌下囊肿的治疗是舌下腺的手术切除。使用OK-432进行硬化疗法是一种公认​​的淋巴畸形治疗方法, 但用于舌下囊肿尚未得到全面评估。

目的：在一项随机双盲试验中, 评估OK-432对舌下囊肿的硬化治疗。

材料和方法：将20例下陷或口内舌下囊肿患者随机分为两组, 分别用OK-432或生理盐水进行双盲注射。研究了对舌下囊肿的影响以及症状和生活质量的评估。

结果：OK-432与安慰剂之间的治疗反应存在显著差异, p¼.041（学生的T检验）。所有口腔内舌下囊肿患者, 以及仅1/4的下陷囊肿患者, 均获得完全反应。注射OK-432后的炎症反应对QOL产生轻度至中度影响。没有观察到严重的并发症。

结论：这项研究表明, OK-432硬化剂治疗囊肿是一种非常有效的治疗口内舌下囊肿的方法, 但对下陷囊肿的作用可能较小。

意义：这是一项有限的研究, 但我们认为, 至少对于口内舌下囊肿, 应建议使用OK-432硬化疗法作为主要治疗方法。

Keywords:

• Ranula
• OK-432
• sclerotherapy
• randomized

Introduction

Ranula is a pseudocyst from the sublingual gland causing a swelling in the floor of the mouth, intraoral ranula, or when the cyst herniates through the mylohyoid muscle a cervical swelling named plunging ranula [1]. It is a rare condition, the prevalence is meagerly studied but in a large cohort study of oral lesions in Sweden Axell et al. reported a prevalence at 0,2/100 [2]. The pathophysiology of ranulas is extravasation of saliva after trauma to the duct or blockage of secretion from a part of the sublingual gland [3,4]. Ranulas are most commonly present in the second or third decade of life. There is a possible genetic impact on the prevalence of plunging ranulas, results from New Zeeland show a ten times higher incidence in Maoris and Polynesians compared to Europeans [5]. Ranulas usually cause moderate discomfort and cosmetic disturbance, but cases with large ranulas causing severe symptoms have been reported [6].

Primary treatment of both oral and plunging ranulas is surgery [1,7]. Different approaches have been used with various results. For intraoral ranulas marsupialization as well as local excision has been used, both with a high rate of recurrence [3,8,9]. In patients with plunging ranulas or in order to effectively treat an intraoral ranula, extirpation of the affected sublingual gland is the treatment of choice [1,3]. This surgical procedure is effective with a low rate of recurrence [3], but requires general anesthesia and could result in complications, such as injury of the lingual nerve. Zhao et al. showed that in 571 patients surgically treated for ranula 5% had sensory deficit on the tongue and 2% damage of Whartons duct [7].

In search of a less invasive treatment option for ranulas sclerotherapy with OK-432 has been suggested. OK-432 is a lyophilized streptococcal preparation originally developed for immunological treatment of cancer. It has been in use for more than 30 years in sclerotherapy of lymphatic malformations [10], and there have been several studies on the efficacy on other benign cystic cervical lesions with favorable results [11–13]. For ranulas a few small studies on OK-432 sclerotherapy have been performed, with various results, ranging from 33% to 87% success rate [13–17]. None of previous studies were randomized trials.

In the Karolinska University Hospital we have previously performed a pilot study where we saw 80% cure rate for ranulas after 1–4 injections with OK-432 (unpublished). No difference was seen between plunging and intraoral ranulas. The procedure was simple and cost-effective and we wanted to study, in a double blinded randomized placebo-controlled trial, if sclerotherapy with OK-432 is an effective and safe method of treatment for ranulas. To our knowledge this has not been done previously.

Material and methods

Study design and eligibility

The trial was approved by the ethics review board at the Karolinska Institutet/University Hospital Stockholm (Approval No 2013/553 31-1, amendment No 2017-1007-32-1) and registered in the EUDRA-CT clinicaltrials.eu (study 2012-004540-29). It was performed in accordance with the ethical standards of the Helsinki Declaration of 1975. Written and oral informed consent was obtained from all patients. The study was performed between November 2013 and September 2019.

The study was conducted at the Karolinska University Hospital, Stockholm, and was designed in collaboration with, and monitored by, Karolinska Trial Alliance (KTA).

The study comprises 20 patients, 10 with plunging and 10 with intraoral ranulas. The cohort consisted of 8 females and 12 males, ages between 9 and 55 years with mean age of 27,8 years. Follow-up from last treatment ranges from 22 to 67 months. Patient characteristics are listed in Table 1.

Table 1. Patient characteristics.

Patient nr	Sex	Age	Side	Intraoral (IO) or plunging (P)	Size ^a,b	Treatment
1	M	33	R	P	40 × 15 × 15	NaCl
2	F	18	R	IO	Moderate	NaCl
3	F	39	R	IO	Moderate	OK-432
4	M	32	L	P	25 × 25 × 10	OK-432
5	M	34	R	P	20 × 7×10	OK-432
6	M	13	R	IO	Moderate	OK-432
7	M	34	R	P	30 × 18 × 10	NaCl
8	M	28	R	P	30 × 30 × 10	NaCl
9	F	12	L	IO	Large	OK-432
10	M	45	R	P	40 × 40 × 15	OK-432
11	M	13	L	IO	Moderate	NaCl
12	F	9	L	IO	Small	NaCl
13	F	31	L	IO	Small	NaCl
14	M	21	R	P	30 × 15 × 20	NaCl
15	M	55	L	P	20 × 18 × 6	OK-432
16	M	51	L	P	40 × 20 × 30	NaCl
17	F	15	R	IO	Large	NaCl
18	M	35	R	P	20 × 10 × 10	NaCl
19	F	14	L	IO	Large	OK-432
20	F	24	L	IO	Moderate	OK

^aPlunging ranulas measured by ultrasound, largest dimension indicated in mm:s.

^bFor intraoral ranulas the size was estimated in tiers; small< 1 cm, moderate 1–2 cm and large >2 cm in largest dimension.

The study was set up as a randomized double-blinded treatment study with two groups; patients given active treatment (OK-432) versus patients given only saline. Based on our pilot study the trial set-up was two injections, approximately 2 months apart. If there was a complete response (CR) already after the first injection only follow-up was performed. The patients were informed that they, if randomized to the placebo-group, got the possibility of treatment with OK-432 after the end of trial, so that all patients included were offered OK-432.

Sample size was based on a power analysis with estimations originating from our pilot study where we saw a cure rate of 80%.

Inclusion and exclusion criteria

Fourty-three patients with ranulas, intraoral or plunging, diagnosed at the ENT-department Karolinska University Hospital, were informed of the trial and assessed for eligibility. For inclusion and exclusion criteria, see Table 2. Twenty-three patients were excluded; main reasons for exclusion were underlying autoimmune disease, patients own wish to get a rapid treatment and not participate in a study, and allergy to penicillin (the preparation of OK-432 contains a small dose of penicillin V to prevent infection).

Table 2. Exclusion and inclusion criteria.

Exclusion criteria:	Inclusion criteria:
Age below 7 years	Correct diagnosis
Allergic to penicillin	No exclusion criteria
Pregnant, breastfeeding or planning to get pregnant during study period	Signed informed consent
Ongoing treatment with anti-inflammatory drugs
Ongoing treatment with immunomodulating drugs
Autoimmune disease

Patient characteristics (age, sex and type of ranula) did not differ substantially between the included and excluded cohorts.

Diagnostic criteria for ranula were for oral ranulas a typical clinical appearance with blue-purple cystic lesion in the floor of the mouth, fine needle aspiration cytology with only mucous secretion and no malignant cells. Diagnosed as having a plunging ranula were patients with typical clinical appearance, ultrasound and/or CT-scan showing a cystic lesion in the submandibular region with no indirect signs of malignancy, FNAC with typical mucous thread- like secretion and cytology with only mucous secretion, and no malignant cells.

Randomization

Patients were randomized by closed numbered envelopes prepared by Karolinska Trial Alliance, managed by a non-blinded study nurse. Key information was stored in a separate room but was available for emergency situations like unexpected side effects.

Intervention

Treatment was given in our outpatient-clinic with no sedatives or other medication. Investigators were three ENT-surgeons with experience of using office-ultrasound in head and neck practice.

OK-432 is not approved for general use in Sweden, but was used after approval on a special license issued by the Swedish Medical Products Agency [18].

The volume of the cyst was approximated based on the ultrasound-investigation for plunging ranulas or clinical examination for intraoral ranulas. The size was noted in the charts, for intraoral ranulas it was assessed as small (<1 cm), moderate (1-2 cm) or large (> 2 cm) in largest dimensions, for plunging ranulas the size in millimeter in three dimensions was measured. Injections were either placebo (saline) or active substance, i.e. OK-432 dissolved in saline. For active injection the dosage was always 0,1 mg OK-432 = 1 IE, in line with the protocol for sclerotherapy with OK-432 in lymphatic malformations at Karolinska University Hospital. The injections, placebo and active substance, were identical in color.

The injection was made by the investigator. For oral ranulas, the injection was made through an oral approach. With the plunging ranulas, an ultrasound-guided puncture was made under sterile conditions. Using connected injection devices (see image 1) the investigator evacuated as much fluid as possible, still preserving the cavity of the ranula, and then injected the prepared solution with or without OK-432 in the same procedure. The total volume of each injection was set by the investigator and aimed to fill up the ranula cavity after evacuation.

Image 1. Injection set-up with the ability to aspirate and inject in one procedure near here

Patients were observed for a minimum of 1 h to check for any immediate severe reaction. They were instructed not to take any anti-inflammatory drugs in order not to disturb the inflammatory response but use paracetamol if needed. All patients had follow-up by telephone and clinical visits. A detailed flow chart of the study is presented in Figure 1.

Figure 1. Flow chart of the study.

Patients with remaining cystic lesions after active treatment with OK-432 were planned for surgery with excision of the sublingual gland. Patients who had received placebo-treatment in the trial were offered treatment with OK-432 according to the same protocol, results are presented below.

Assessments and evaluation

Response to treatment was assessed by the investigator with clinical examination including office-based ultrasound. The patients with intraoral ranulas were also examined with neck-ultrasound to ensure that no deep part of remaining ranula was overlooked.

Complete response (CR) was defined as no sign of any remaining cystic lesion, regardless of size.

Patients received two evaluation forms regarding symptoms after the injection; a diary to post inflammatory symptoms (not validated), and EQ5D-5L Sweden QOL protocol. The EQ5D-5L protocol is a European standardized and validated tool to assess health outcomes [19].

The costs per patient for surgery (excision of the sublingual gland) compared to one series of injections and control was calculated with help from a hospital economic controller.

Continuously the trial was monitored by Karolinska Trial Alliance.

To compare treatment response between the two groups we used a two-tailed student’s t-test.

Results

Primary outcome was the response to treatment assessed by clinical examination and/or office ultrasound examination. Patients were evaluated for a total period ranging from 16- 28 months.

19 out of 20 patients completed the trial. The one drop-out was due to the only adverse event noted: an infection after injection of placebo.

No serious adverse events were noted.

Six of the 9 patients treated with OK-432 were responders. Two were complete responders already after one injection, the other 4 after the second treatment. Of the 10 patients treated with placebo, 2 were responders and 8 non-responders. Both patients cured by placebo had intraoral ranulas (Table 3). Treatment response differed significantly between the OK-432 group and the placebo group, with a p = .041 on a two-tailed student’s T-test assuming equal variance.

In the responder-group treated with OK-432 5 had oral ranulas and 1 plunging. All 3 non-responders to OK-432-treatment had plunging ranulas.

To motivate patients to participate in the trial we made a study design where the patients treated with placebo were offered OK-432-sclerotherapy in a second round, out of trial. Eight patients got treated in this group, i.e. the whole placebo-group except the 2 patients who responded to placebo. Of these 8 patients, 5 had plunging ranulas and three had oral. Only 1 of the plunging ranulas in this group responded, compared to the oral group were all 3 had a complete response after OK-432 treatment.

If we summarize both the randomized and non-randomized patients a total of 8 oral and 9 plunging ranulas were treated with 1 or 2 injections with OK-432. Nine were treated in a double-blinded randomized setting, the other 8 in a non-blinded second round. All the oral ranulas were complete responders but only 2 (out of 9) of the plunging.

Secondary outcome was the level of discomfort due to inflammatory reaction after sclerotherapy, evaluated by EQ-5D and patient diary. 12/20 patients enrolled in the study handed in these data.

The discomfort and impact on QOL after double-blinded sclerotherapy with OK-432 varied, with a change in EQ-5D-5L-index from pre-treatment to the worst day of inflammatory reaction between − 0,44 to −0,21 (Table 4). The patients who were treated with OK-432 in a non-blinded setting (after placebo) seemed to have a worse experience, with a change in EQ-5D-5L-index up to − 0,79, see Table 4.

Table 3. Summary of results in the randomized groups.

Double-blinded randomized treatment:
	Responder	Non responder
OK-432	6	3
Placebo	2	8

Table 4. Changes in EQ-5D-5L score peak symptomatic day vs before treatment, and summary of patient diary, from the patients in the randomized groups who returned forms.

Inflammatory symptoms score and diary, randomised trial:
Patient nr	Baseline score	Lowest score	Difference	Days w sym	Bodytemp max	Treatment
1	1	1	0		37.2	NaCL
11	1	0.8366	−0.1634	2	37.1	NaCl
13	1	1	0		37.2	NaCL
14	1	1	0		37.2	NaCl
16	1	0.8366	−0.1634	5		NaCL
17	1	1	0			NaCl
18	1	1	0			NaCl
8	1	0.767	−0.233	2		NaCl^a
3	1	0.796	−0.204	9	38.2	OK-432
4	1	0.5603	−0.4397	5	38	OK-432
6	1	0.76	−0.24	5		OK-432
19	1	0.7397	−0.2603	3		OK-432

^aDrop-out of study due to local infection in ranula after injection.

From the patient diaries we noted that the inflammatory reaction lasted 2-9 days, (mean 5,3 days) with predominant symptoms of swelling, pain, fever up to 38,7 degrees. From the different dimensions in EQ-5D-5L we see very little impact on mobility or self-care, low impact on usual activities and anxiety/depression, and moderate impact on pain/discomfort.

Patients treated with saline showed mild symptoms or effect on QOL, except the patient who got an infection after his injection and dropped out of the trial.

In addition to these results, we have performed an approximative comparison of the costs of sublingual gland excision versus injection of OK-432. In this we included costs for staff, material, day-surgery bed and drug related costs. The results show that in the Karolinska University Hospital the cost for surgery i.e. sublingual gland excision is ten times higher than the procedure with two injections of OK-432.

Discussion

Our study suggests that sclerotherapy with OK-432 is a safe treatment with a very good efficacy for patients with intraoral ranulas but possibly less useful in plunging ranulas. We can also conclude that the inflammatory reaction is well tolerable.

To our knowledge this is the first study to evaluate the effect of sclerotherapy with OK-432 on ranulas in a randomized, double-blinded placebo-controlled setting. The small sample size limits the conclusions to be made, but still some important findings can be highlighted.

The results show that 6 out of 9 patients in the group primarily randomized to OK-432 were complete responders, compared to 2 out of 10 in the placebo-group. In the OK-432- group all patients with intraoral ranulas were complete responders, but only 1 out of 4 with plunging ranulas. If we also include the treatment of the non-blinded patients, who initially got placebo, we see that in total all of the oral ranulas (9) were complete responders to the treatment, but only 2 out of 9 of the plunging. Our results thus suggest that there is a considerable difference in response to sclerotherapy with OK432 between the oral and plunging ranulas.

The effect of sclerotherapy OK-432 on intraoral ranulas is consistent with the findings of previous studies OK-432. For plunging ranulas previous results have been inconsistent, however both our pilot study and other investigators have reported good results also for plunging ranulas [14,17]. The plunging ranulas could have a more complex wall-structure given the path through the mylohyoid muscle, and therefore it could be more difficult to reach the whole cyst with injection. Another possibility is that the number of injections in this protocol is not sufficient, or that plunging ranulas may need a higher total dose of OK-432. A different hypothesis is that the effect could be mediated by inflammation of the sublingual gland itself, not in the cystic walls. This could explain why oral ranulas respond so well, since the cystic lesion in that case is right adjacent to the gland. Supporting this theory is a study by Murakami et al., where percutaneous chemical ablation of the sublingual gland with a combination of sodium tetradecyl sulfate and alcohol showed at cure rate of 87,5% in 24 patients with ranulas, both intraoral and plunging [20]. If this explanation holds true, a more effective way of treating plunging ranulas would be to inject OK-432 directly into the sublingual gland.

Notably two patients with intraoral ranulas randomized to placebo were cured. A possible explanation could be either that even a very mild irritation caused by our saline injection can start an inflammatory response or that ranulas in rare cases may recover spontaneously.

We did not experience any major concern regarding the safety using OK-432.The only adverse event noted was an infection related to injection of NaCl, showing the importance of sterile conditions at the time of the injection.

We found the inflammatory response in this study is in line with previous reports [14–16]. Our conclusion is that the inflammatory response after sclerotherapy with OK-432 is tolerable and does not cause any serious side effects. However, the inflammatory response was very heterogenous in duration and severity. One patient experienced considerable impact on all aspects of QOL, indicating that for some patients the inflammatory reaction may cause great but temporary discomfort.

To evaluate the benefit of sclerotherapy compared to surgical removal of the sublingual gland there are several aspects to consider. The main issue is to put the efficacy and the discomfort after sclerotherapy in relation to the perioperative risks and postoperative pain and other possible surgical side effects related to extirpation of the sublingual gland. To our knowledge no similar evaluation of QOL and postoperative symptoms after excision of the sublingual gland has been made, but such a study would give valuable input.

A clear advantage of sclerotherapy is that it demands very little resources from the health-care provider; it can be performed as a simple outpatient procedure, with no need for anesthesia, operation room or hospital bed. In our study the costs for surgery are approximately ten times higher than sclerotherapy with two injections.

Methodological considerations/limitations

The strength of this trial is the double-blinded randomized setting and the long follow-up time. The weakness is the small sample size. Ranula is a quite rare condition and the fact that it is hard to get large cohorts of patients is commented in several reviews [1,2]. This entails that larger randomized trials are difficult to transact.

Conclusion

In summary, the results of our study confirm the conclusion that OK-432 is an effective treatment for patients with intraoral ranulas, but possibly less effective in patients with plunging ranulas. To further compare surgical treatment with sclerotherapy it would be useful with comparable data on QOL and discomfort after extirpation of the sublingual gland. Sclerotherapy with OK-432 is a minimal-invasive, cost-effective treatment with no known long-term complications contrary to surgery. We thus suggest that sclerotherapy should be established as an alternative primary treatment, at least for patients with intraoral ranulas.

Disclosure statement

No potential conflict of interest was reported by the author(s).