- Email: [email protected]
Doxycycline sclerotherapy for pediatric head and neck macrocystic lymphatic malformations: A case series and review of the literature
International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
Contents lists available at SciVerse ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Doxycycline sclerotherapy for pediatric head and neck macrocystic lymphatic malformations: A case series and review of the literature Nausheen Jamal a,b,*, Sameer Ahmed b, Todd Miller c, John Bent a, Allan Brook c, Sanjay Parikh a,d, Ashish Ankola a a
Department of Otorhinolaryngology-Head and Neck Surgery, Albert Einstein College of Medicine, 3400 Bainbridge Ave, 3rd Floor, Bronx, NY 10467, United States Department of Head and Neck Surgery, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave., CHS 62-132, Los Angeles, CA 90095-1624, United States Department of Radiology, Division of Neuroradiology, Interventional Neuroradiology Section, Albert Einstein College of Medicine, Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467, United States d Division of Otolaryngology, Seattle Children’s Hospital, W-7729 – Otolaryngology-ENT, 4800 Sand Point Way NE, Seattle, WA 98105, United States b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 22 January 2012 Received in revised form 9 April 2012 Accepted 14 April 2012 Available online 7 May 2012
Objective: (a) To evaluate the efficacy of doxycycline as a percutaneous sclerotherapy agent in pediatric head and neck macrocystic lymphatic malformations (LM) and (b) to review the literature with regard to recent developments in the treatment of lymphatic malformations using sclerotherapy. Methods: We reviewed the medical records and imaging studies of all patients who underwent percutaneous sclerotherapy of macrocystic LM of the head and neck at our institution between June 2005 and May 2010. All studies were reviewed and procedures performed by a single interventional neuroradiologist using computed tomography (CT) guidance. LM were individually cannulated, the contents aspirated, and then injected with doxycycline at concentrations of 10–20 mg/ml. Response to sclerotherapy was determined clinically. Results: Seven patients underwent a total of eight sclerotherapy treatments during the study period. Of the six patients with appropriate follow-up, 67% have experienced complete or near-complete clinical resolution of their LM (1243 days mean follow-up), while 33% have developed recurrent swelling after an initial response following a single doxycycline injection (53 days mean follow-up). Conclusions: Our institutional results, in combination with recently published findings, support the moderate efficacy and excellent safety profile of percutaneous doxycycline sclerotherapy for macrocystic lymphatic malformations. ß 2012 Elsevier Ireland Ltd. All rights reserved.
Keywords: Lymphatic malformation Lymphangioma Doxycycline Sclerotherapy
1. Introduction Lymphatic malformations (LM) are benign tumors that are the result of abnormal embryogenesis of the lymphatic system [1]. They are common among vascular malformations in the pediatric population [2], and over 50% occur within the head and neck region [3,4]. These lesions are subdivided into three types: macrocystic lesions, which contain one or more cysts that are at least 2 cm3 in volume; microcystic lesions, which contain cysts that are individually less than 2 cm3 in volume; and mixed lesions, which contain both macrocysts and microcysts [4–6]. Due to their location in the head and neck, LM may cause significant functional and cosmetic morbidity. Although spontaneous resolution of these tumors may occur, the reported rate is
* Corresponding author at: Division of Head and Neck Surgery, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave., CHS 62-132, Los Angeles, CA 90095-1624, United States. Tel.: +1 310 825 5179; fax: +1 310 825 2810. E-mail address: [email protected] (N. Jamal). 0165-5876/$ – see front matter ß 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijporl.2012.04.015
disappointingly low at 3% [1,4]. As such, early treatment provides symptomatic, functional, and cosmetic improvement in otherwise persistent lesions [7]. Traditionally, surgery has been considered the standard of care for LM [1,8], but results are inconsistent. In 1995, de Serres et al. published a staging system that helped predict the results of extirpation based on lesion location. In this article, infrahyoid location and unilateral disease were correlated with improved outcomes. Surgery, however, was not without significant morbidity. They reported numerous complications, including cranial nerve injury, malocclusion, and poor cosmetic results; the rate of complications increased from 17% in stage I lesions to 100% in stage V lesions [9]. Molitch reported on 5 patients with lymphatic malformations in different parts of the body that were treated with percutaneous sclerotherapy using doxycycline, a broad-spectrum antibiotic from the tetracycline class, with size reduction or symptomatic resolution noted in all patients [10]. Since that time, numerous sclerosants have been reported in the literature to treat lymphatic malformations (refer to Table 1). Ethanol, sodium tetradecyl sulfate (STS), bleomycin, and OK-432 (Picibanil) have all been
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
1128
Table 1 Sclerosing agents used in the treatment of lymphatic malformations.
(Figs. 2 and 3). This was followed by an injection of doxycycline reconstituted in sterile water at a concentration of 10–20 mg/ml to fill the cavity (the highest doxycycline dose delivered in a single treatment was 233 mg). Nearly all patients (5 out of 7) were treated with peri-operative antibiotics with gram-positive coverage. No drains were placed. No immediate complications were noted. Procedures were performed on an outpatient basis, unless the patient was already hospitalized at the time of procedure.
Doxycycline Bleomycin OK-432 Ethanol Sodium tetradecyl sulfate (STS) Ethibloc Acetic acid
employed in treatment of head and neck LM based on experiences with other sclerotherapy procedures, and with varying but generally positive results [2,6,7,11]. These positive results have been consistent among macrocystic lesions; the results with microcystic lesions among the various sclerosants have been mixed [2,8,12,13]. For lesions refractory to sclerotherapy, surgery can be used as a salvage option; anecdotally, the use of sclerosants does not cause increased scarring and consequently does not make surgery more difficult [8]. Of the various sclerosants, doxycycline is gaining popularity, due to its low cost, ease of availability – particularly outside of research protocols – and excellent safety and efficacy profiles [4,13–15]. To verify the efficacy of doxycycline sclerotherapy, we conducted (a) a retrospective analysis of our experience with primary doxycycline sclerotherapy for pediatric head and neck macrocystic LM as well as (b) a review of the recent relevant sclerotherapy literature. 2. Materials and methods Medical records and imaging studies of all patients who underwent percutaneous sclerotherapy of macrocystic LM of the head and neck between June 2005 and May 2010 were reviewed. Overall, seven patients presented during the study period and underwent eight sclerotherapy injections; 43% of the patients injected were female and 57% were male. Patients ranged in age from 1 month to 10 years at the time of treatment. Post-treatment follow-up for all patients averaged 725 days and ranged from 32 to 1511 days, with 2 patients lost to follow-up (Table 2). All imaging studies were reviewed by a single interventional radiologist, who determined that the radiologic findings in all cases were consistent with a macrocystic lymphatic malformation. Most procedures (6 out of 8) were performed under general anesthesia. The fluid-filled space was cannulated using a small gauge needle under CT-guidance. Cystic fluid contents were aspirated and sporadically sent for cytologic and/or microbiologic analyses. 1–3 cm3 of contrast material (Omnipaque 180, Amersham Health, Princeton, NJ) was then instilled to outline the cavity and determine any continuity with adjacent structures
3. Results Of the seven patients in our study, one was lost to follow-up after his doxycycline treatment session. Of the six remaining patients who had at least one follow-up visit, two have undergone complete resolution of their LM (Figs. 1 and 4) without further incident (992 days mean follow-up) following an average of 1.5 treatments; the four remaining patients developed recurrent swelling. For three of the four patients with recurrent swelling, the recurrences coincided with episodes of infection. One of these three patients experienced swelling 4 years after sclerotherapy, which resolved without further incident following conservative treatment with oral antibiotics (now at 1477 days follow-up). Another of these three achieved near complete resolution, but he continues to demonstrate mild swelling of the right neck. He rarely develops infections in this area that result in increased swelling, but these resolve quickly with oral antibiotics. Because of the rarity of infection and the subtle nature of the swelling present at baseline, the family has opted to pursue conservative management only (now at 1511 days of follow-up). The last of the three patients – with swelling that was related to infection – required hospitalization for treatment with intravenous antibiotics. Although the swelling has since improved, the family is considering a second treatment with sclerotherapy (at 74 days follow-up). The patient who developed a recurrence of a fluid collection without an infectious cause (Patient #1 in Table 2) was available for only one follow-up visit at 32 days. This child is a part of the foster care system which has impaired our ability to regularly follow up with her. With regard to complications, one of our patients presented to the emergency room after developing a cellulitis which required admission to the hospital for intravenous antibiotics. He responded well and was soon discharged on an oral antibiotic course. A different patient complained of pain at the time of procedure; he was one of only two patients who did not receive general anesthesia. Notably, after appropriate intra-operative pain control, his post-operative pain was unremarkable. None of the patients reported complications such as fevers, chills, or post-operative pain.
Table 2 Clinical presentations, outcomes, and follow-up. Age listed is age at the time of initial treatment. Age/sex
Lesion size and location
Stage
Number of treatments
Peri-procedural antibiotics
Clinical results
Follow-up (days)
1 2 3
3y/F 10y/M 2m/M
II II IV
1 1 1
None Amoxicillin/clavulanate Clindamycin
Recurrence Unknown Complete resolution
32 0 1115
4
7y/F
Right neck, level II: 8 cm 5 cm 6 cm Right parotid: 4 cm 3 cm 3.5 cm Right submandibular space: 3.9 cm 3.4 cm 3 cm Left parotid and submandibular space: 4 cm 4.5 cm 4 cm
II
1
Cefazolin
1477
5
1m/F
Right mandible: 6 cm 2 cm 5 cm
IV
1
Ampicillin/sulbactam
6 7
2y/M 3y/M
Left parotid tail: 4 cm 6 cm 2.5 cm Right neck: 7 cm 6 cm 4 cm
II II
2 1
Cephalexin None
Developed swelling with infection, now with complete resolution following conservative treatment Recurrence with infection, plan for 2nd sclerotherapy treatment Complete resolution Near complete resolution, develops mild infection rarely that responds to oral antibiotics
74 868 1511
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
1129
Fig. 3. Image capture during repeat sclerotherapy procedure showing persistent but smaller lesion.
In summary, 50% of our patients with at least one follow-up visit had complete resolution of their lymphatic malformation. When the complete and near-complete resolution patients were combined, our rate of successful treatment was 67% (at 1243 days mean follow-up). In other words, 33% of our patients did not experience significant clinical improvement of their LM or developed a recurrence at the endpoint of the study (follow-up time for these patients was 53 days).
4. Discussion
Fig. 1. Photograph depicting a typical pre-treatment clinical presentation.
Fig. 2. Image capture from pictured patient during initial CT-guided sclerotherapy procedure.
Although doxycycline has been used as a sclerosant for years, its exact mechanism of action in this capacity remains unclear. It appears to induce an inflammatory reaction within the endothelial-lined cavity that leads to deposition of fibrin and collagen with eventual involution [14,15]. It also acts as an angiogenesis inhibitor by interfering with cell proliferation and migration. It does so by inhibiting matrix metalloproteinase (MMP) and suppressing vascular endothelial growth factor (VEGF)-induced angiogenesis and lymphangiogenesis [14]. Its efficacy has been previously established, and our rates of significant LM resolution are comparable with the rates of 70–100% found in the literature [7,10,13,14]. Interestingly, those patients in our series with complete resolution have the greatest length of follow-up, suggesting that use of doxycycline sclerotherapy correlates with improved results in the long-term. However, our results indicate that not all macrocystic LM respond to doxycycline sclerotherapy as one-third of our patients did not demonstrate significant clinical improvement. Complications noted in other studies, such as post-treatment hemorrhage, fever, scarring, and development of Horner’s syndrome [7,10,14,15] did not occur in our patients, although one child in our series was hospitalized for an infection requiring intravenous antibiotics. Although post-treatment systemic levels of doxycycline are elevated at supra-therapeutic levels in patients who receive a standard sclerotherapy dose (median level = 16 mg/ ml, compared to a normal therapeutic range of 1.5–2.1 mg/ml), no drug-induced side effects are noted in these patients [14]. The most significant adverse effect associated with tetracyclines is the irreversible staining of permanent teeth and bone. Doxycycline, however, has less affinity for bone and its association with dental staining is insignificant [7].
1130
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
Fig. 4. Post-treatment photograph of previously pictured patient.
Doxycycline has been reported to cause severe discomfort on injection, requiring general anesthesia in most patients [14]. This is consistent with our experience. One of the two procedures performed using local anesthesia with sedation was interrupted by patient complaints of significant pain, requiring several doses of narcotic pain medication. Post-procedure, however, the patient’s pain was easily controlled with over-the-counter analgesics. As noted in Table 1, multiple sclerosants are currently employed to treat lymphatic malformations. Bleomycin is one such drug that initially gained popularity as a sclerosing agent due to its known efficacy in the treatment of pleural effusions. Preliminary results were promising, with no major complications. One series, however, despite LM resolution rates of about 67%, led to a treatment-related death [7,16]. Its use further demands caution due to the well-established risk of pulmonary toxicity at doses greater than 500 mg [11]. OK-432 is a biologic preparation of lyophilized powder containing Streptococcus pyogenes Su strain cells (group A, type 3) treated with benzylpenicillin G that was popularized in Japan [7,11,17]. Like doxycycline, its mechanism of action is also unclear, although it appears to alter the white blood cell population, thereby inducing production of various cytokines. One such cytokine, tumor necrosis factor, increases endothelial permeability, thus allowing for egress of lymphatic fluid with resultant cell death [7]. In the United States, use of OK-432 is restricted to research protocols. Studies have shown it to be effective in treatment of macrocystic LM, although its efficacy appears limited in both microcystic lesions as well as post-surgical ones [7,18]. Reactions to treatment with OK-432 include fever, anorexia,
malaise, nausea, and hypersensitivity. Its use is contraindicated in patients with a penicillin allergy. Ethanol is a potent and widely used sclerosing agent. Its mechanism of action is via rapid cellular dehydration with precipitation of protoplasm as it comes into contact with the endothelial lining of lymphatic malformations [7]. However, it yields inconsistent and unreliable treatment outcomes, and its use is limited – particularly in small children with large lesions – due to the risk of systemic toxicity [2,14]. At doses of 1 ml/kg, patients may have increased serum ethanol levels that put them at risk for respiratory depression, cardiac arrhythmias, rhabdomyolysis, hypoglycemia, and seizures [7]. Furthermore, it cannot be injected near major nerve trunks due to its neurolytic effects, which further limits its use in the head and neck region [7,14]. Hu et al. recently found that ethanol sclerotherapy for facial venous malformations predisposes the zygomatic and temporal branches of the facial nerve to injury [19]. Sodium tetradecyl sulfate (STS), a detergent, is one of the most common sclerosing agents used; it is commercially available for this purpose. It is reported to cause less swelling than ethanol and is uniquely used for injection of orbital LM. In addition to a possible association with increased rates of infection, at least one episode of vision loss has been reported [7]. A recent case series by Shiels et al. reported on the efficacy of multi-drug percutaneous sclerotherapy. After individual aspiration of cysts, macrocystic LM were pre-treated for 2 min with 3% STS, followed by a 15 min treatment with 98% ethanol. Microcysts were also individually aspirated and then injected with doxycycline. Remarkably, 93% of 28 macrocyst patients required only one treatment session to attain complete resolution of their LM, with 100% resolution of all individually treated cysts (including both macrocysts and microcysts) within 3 days of completing treatment. The authors hypothesize that their excellent result in patients undergoing fewer than average treatments, specifically in macrocysts, is the direct result of combination therapy. They postulate that STS functions as a detergent to release transmembrane lipoproteins and increase LM membrane permeability, allowing greater penetration of ethanol, the latter of which causes intracellular protein denaturation, leading to cell death. Perhaps the most significant finding in this study was the successful treatment of microcystic lesions, albeit only those cysts large enough to be individually cannulated, aspirated, and injected with doxycycline [2]. However, microcystic and mixed LM continue to be challenging problems. Cahill et al. recently reviewed their experience with doxycycline sclerotherapy for LM in neonates and infants. For the ten patients with macrocystic LM, doxycycline at a concentration of 10 mg/ml was catheter instilled while for the seven patients with mixed LM, doxycycline at a concentration of 10 mg/ml or STS foam was directly injected into the microcyst. The majority of the mixed LM patients (6/7) were treated with a direct injection of doxycycline as opposed to a direct injection with STS foam. Their data demonstrated two-thirds of patients with macrocystic LM had significant improvement, in which a large subset of this group had near total resolution. However, less than 50% of patients with mixed LM treated with doxycycline injection noted significant improvement [13]. Our study certainly has weaknesses: it is a retrospective review with low power. Nevertheless, our results, in combination with a review of the literature, demonstrate moderate efficacy for doxycycline as a single sclerotherapeutic modality for pediatric macrocystic head and neck LM as well as the possibility of increased benefit with carefully chosen combination sclerotherapy in mixed/microcystic LM. Another potential weakness is the variable concentration of doxycycline in our study (ranging between 10 mg/ml and 20 mg/
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
ml). Although different concentrations were used based on the volume of the macrocyst being treated, the total dose of doxycycline never exceeded 233 mg per treatment session. In addition, both ends of the range are within the limits of acceptable treatment protocols. For example, Shiels et al. used a concentration of 20 mg/ml in their study while Cahill et al. and Burrows et al. used a concentration of 10 mg/ml in their studies [2,13,14]. The relative ease and efficacy of treatment combined with the relatively low rate of treatment-related morbidity are strong points that support the superiority of percutaneous sclerotherapy as an alternative to traditional surgical extirpation. However, the sclerotherapy associated side effects cannot be entirely dismissed, as more research is shedding light onto which complications occur most frequently and which patient groups are at higher risk (i.e. infants and neonates) [13]. Interestingly, our data suggest that the rate of resolution is directly proportional to the length of followup, implying that the full therapeutic process may have not yet occurred in our patients with incomplete resolution; indeed, these patients are also the individuals with the shortest length of posttreatment follow-up. What remains to be determined is the appropriate selection, combination, and dosing regimens of specific sclerotherapeutic agents, and the specific time course in which resolution should be expected to occur for both macrocystic and microcystic/mixed LM. Conflict of interest statement The authors report no conflicts of interest in this study. References [1] R.J. Smith, Lymphatic malformations, Lymphat. Res. Biol. 2 (1) (2004) 25–31. [2] W.E. Shiels, D.R. Kang, J.W. Murakami, M.J. Hogan, G.J. Wiet, Percutaneous treatment of lymphatic malformations, Otolaryngol. Head Neck Surg. 141 (2) (2009) 219–224. [3] T.L. Kennedy, M. Whitaker, P. Pellitteri, W.E. Wood, Cystic hygroma/lymphangioma: a rational approach to management, Laryngoscope 111 (2001) 1929– 1937.
1131
[4] D. Nehra, L. Jacobson, P. Barnes, B. Mallory, C.T. Albanese, K.G. Sylvester, Doxycycline sclerotherapy as primary treatment of head and neck lymphatic malformations in children, J. Pediatr. Surg. 43 (3) (2008) 451–460. [5] C.M. Giguere, N.M. Bauman, Y. Sato, D.K. Burke, J.H. Greinwald, S. Pransky, et al., Treatment of lymphangiomas with OK-432 (Picibanil) sclerotherapy: a prospective multi-institutional trial, Arch. Otolaryngol. Head Neck Surg. 128 (10) (2002) 1137–1144. [6] M.C. Smith, M.B. Zimmerman, D.K. Burke, N.M. Bauman, Y. Sato, R.J. Smith, Efficacy and safety of OK-432 immunotherapy of lymphatic malformations, Laryngoscope 119 (1) (2009) 107–115. [7] A.I. Alomari, V.E. Karian, D.J. Lord, H.M. Padua, P.E. Burrows, Percutaneous sclerotherapy for lymphatic malformations: a retrospective analysis of patientevaluated improvement, J. Vasc. Interv. Radiol. 17 (2006) 1639–1648. [8] J.L. Acevedo, R.K. Shah, S.E. Brietzke, Nonsurgical therapies for lymphangiomas: a systematic review, Otolaryngol. Head Neck Surg. 138 (April (4)) (2008) 418– 424. [9] L.M. de Serres, K.C. Sie, M.A. Richardson, Lymphatic malformations of the head and neck. A proposal for staging, Arch. Otolaryngol. Head Neck Surg. 121 (5) (1995) 577–582. [10] H.I. Molitch, E.C. Unger, C.L. Witte, E. vanSonnenberg, Percutaneous sclerotherapy of lymphangiomas, Radiology 194 (2) (1995) 343–347. [11] D.C. Bloom, J.A. Perkins, S.C. Manning, Management of lymphatic malformations, Curr. Opin. Otolaryngol. Head Neck Surg. 12 (2004) 500–504. [12] S. Wiegand, B. Eivazi, A.P. Zimmermann, A.M. Sesterhenn, J.A. Werner, Sclerotherapy of lymphangiomas of the head and neck, Head Neck 33 (November (11)) (2011) 1649–1655. [13] A.M. Cahill, E. Nijs, D. Ballah, D. Rabinowitz, L. Thompson, N. Rintoul, et al., Percutaneous sclerotherapy in neonatal and infant head and neck lymphatic malformations: a single center experience, J. Pediatr. Surg. 46 (November (11)) (2011) 2083–2095. [14] P.E. Burrows, R.K. Mitri, A. Alomari, H.M. Padua, D.J. Lord, M.B. Sylvia, et al., Percutaneous sclerotherapy of lymphatic malformations with doxycycline, Lymphat. Res. Biol. 6 (3–4) (2008) 209–216. [15] B.M. Cordes, F.G. Seidel, M. Sulek, C.M. Giannoni, E.M. Friedman, Doxycycline sclerotherapy as the primary treatment for head and neck lymphatic malformations, Otolaryngol. Head Neck Surg. 137 (6) (2007) 962–964. [16] K.H. Kim, M.W. Sung, J.L. Roh, M.H. Han, Sclerotherapy for congenital lesions in the head and neck, Otolaryngol. Head Neck Surg. 131 (2004) 307–316. [17] W. Mol, H. Furukawa, S. Sasaki, U. Tomaru, T. Hayashi, A. Saito, Evaluation of the sclerotherapeutic efficacy of ethanol, polidocanol, and OK-432 using an in vitro model, Dermatol. Surg. 33 (December (12)) (2007) 1452–1459. [18] D.K. Gupta, M. Bajpai, M. Srinivas, A. Lall, S. Dave, S. Ogita, Efficacy of OK-432 in pediatric lymphangiomas: preliminary experience, J. Indian Assoc. Pediatr. Surg. 4 (1999) 192–195. [19] X. Hu, D. Chen, C. Jiang, Y. Jin, H. Chen, G. Ma, et al., Retrospective analysis of facial paralysis caused by ethanol sclerotherapy for facial venous malformation, Head Neck 33 (November (11)) (2011) 1616–1621.
Contents lists available at SciVerse ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Doxycycline sclerotherapy for pediatric head and neck macrocystic lymphatic malformations: A case series and review of the literature Nausheen Jamal a,b,*, Sameer Ahmed b, Todd Miller c, John Bent a, Allan Brook c, Sanjay Parikh a,d, Ashish Ankola a a
Department of Otorhinolaryngology-Head and Neck Surgery, Albert Einstein College of Medicine, 3400 Bainbridge Ave, 3rd Floor, Bronx, NY 10467, United States Department of Head and Neck Surgery, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave., CHS 62-132, Los Angeles, CA 90095-1624, United States Department of Radiology, Division of Neuroradiology, Interventional Neuroradiology Section, Albert Einstein College of Medicine, Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467, United States d Division of Otolaryngology, Seattle Children’s Hospital, W-7729 – Otolaryngology-ENT, 4800 Sand Point Way NE, Seattle, WA 98105, United States b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 22 January 2012 Received in revised form 9 April 2012 Accepted 14 April 2012 Available online 7 May 2012
Objective: (a) To evaluate the efficacy of doxycycline as a percutaneous sclerotherapy agent in pediatric head and neck macrocystic lymphatic malformations (LM) and (b) to review the literature with regard to recent developments in the treatment of lymphatic malformations using sclerotherapy. Methods: We reviewed the medical records and imaging studies of all patients who underwent percutaneous sclerotherapy of macrocystic LM of the head and neck at our institution between June 2005 and May 2010. All studies were reviewed and procedures performed by a single interventional neuroradiologist using computed tomography (CT) guidance. LM were individually cannulated, the contents aspirated, and then injected with doxycycline at concentrations of 10–20 mg/ml. Response to sclerotherapy was determined clinically. Results: Seven patients underwent a total of eight sclerotherapy treatments during the study period. Of the six patients with appropriate follow-up, 67% have experienced complete or near-complete clinical resolution of their LM (1243 days mean follow-up), while 33% have developed recurrent swelling after an initial response following a single doxycycline injection (53 days mean follow-up). Conclusions: Our institutional results, in combination with recently published findings, support the moderate efficacy and excellent safety profile of percutaneous doxycycline sclerotherapy for macrocystic lymphatic malformations. ß 2012 Elsevier Ireland Ltd. All rights reserved.
Keywords: Lymphatic malformation Lymphangioma Doxycycline Sclerotherapy
1. Introduction Lymphatic malformations (LM) are benign tumors that are the result of abnormal embryogenesis of the lymphatic system [1]. They are common among vascular malformations in the pediatric population [2], and over 50% occur within the head and neck region [3,4]. These lesions are subdivided into three types: macrocystic lesions, which contain one or more cysts that are at least 2 cm3 in volume; microcystic lesions, which contain cysts that are individually less than 2 cm3 in volume; and mixed lesions, which contain both macrocysts and microcysts [4–6]. Due to their location in the head and neck, LM may cause significant functional and cosmetic morbidity. Although spontaneous resolution of these tumors may occur, the reported rate is
* Corresponding author at: Division of Head and Neck Surgery, David Geffen School of Medicine at UCLA, 10833 Le Conte Ave., CHS 62-132, Los Angeles, CA 90095-1624, United States. Tel.: +1 310 825 5179; fax: +1 310 825 2810. E-mail address: [email protected] (N. Jamal). 0165-5876/$ – see front matter ß 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijporl.2012.04.015
disappointingly low at 3% [1,4]. As such, early treatment provides symptomatic, functional, and cosmetic improvement in otherwise persistent lesions [7]. Traditionally, surgery has been considered the standard of care for LM [1,8], but results are inconsistent. In 1995, de Serres et al. published a staging system that helped predict the results of extirpation based on lesion location. In this article, infrahyoid location and unilateral disease were correlated with improved outcomes. Surgery, however, was not without significant morbidity. They reported numerous complications, including cranial nerve injury, malocclusion, and poor cosmetic results; the rate of complications increased from 17% in stage I lesions to 100% in stage V lesions [9]. Molitch reported on 5 patients with lymphatic malformations in different parts of the body that were treated with percutaneous sclerotherapy using doxycycline, a broad-spectrum antibiotic from the tetracycline class, with size reduction or symptomatic resolution noted in all patients [10]. Since that time, numerous sclerosants have been reported in the literature to treat lymphatic malformations (refer to Table 1). Ethanol, sodium tetradecyl sulfate (STS), bleomycin, and OK-432 (Picibanil) have all been
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
1128
Table 1 Sclerosing agents used in the treatment of lymphatic malformations.
(Figs. 2 and 3). This was followed by an injection of doxycycline reconstituted in sterile water at a concentration of 10–20 mg/ml to fill the cavity (the highest doxycycline dose delivered in a single treatment was 233 mg). Nearly all patients (5 out of 7) were treated with peri-operative antibiotics with gram-positive coverage. No drains were placed. No immediate complications were noted. Procedures were performed on an outpatient basis, unless the patient was already hospitalized at the time of procedure.
Doxycycline Bleomycin OK-432 Ethanol Sodium tetradecyl sulfate (STS) Ethibloc Acetic acid
employed in treatment of head and neck LM based on experiences with other sclerotherapy procedures, and with varying but generally positive results [2,6,7,11]. These positive results have been consistent among macrocystic lesions; the results with microcystic lesions among the various sclerosants have been mixed [2,8,12,13]. For lesions refractory to sclerotherapy, surgery can be used as a salvage option; anecdotally, the use of sclerosants does not cause increased scarring and consequently does not make surgery more difficult [8]. Of the various sclerosants, doxycycline is gaining popularity, due to its low cost, ease of availability – particularly outside of research protocols – and excellent safety and efficacy profiles [4,13–15]. To verify the efficacy of doxycycline sclerotherapy, we conducted (a) a retrospective analysis of our experience with primary doxycycline sclerotherapy for pediatric head and neck macrocystic LM as well as (b) a review of the recent relevant sclerotherapy literature. 2. Materials and methods Medical records and imaging studies of all patients who underwent percutaneous sclerotherapy of macrocystic LM of the head and neck between June 2005 and May 2010 were reviewed. Overall, seven patients presented during the study period and underwent eight sclerotherapy injections; 43% of the patients injected were female and 57% were male. Patients ranged in age from 1 month to 10 years at the time of treatment. Post-treatment follow-up for all patients averaged 725 days and ranged from 32 to 1511 days, with 2 patients lost to follow-up (Table 2). All imaging studies were reviewed by a single interventional radiologist, who determined that the radiologic findings in all cases were consistent with a macrocystic lymphatic malformation. Most procedures (6 out of 8) were performed under general anesthesia. The fluid-filled space was cannulated using a small gauge needle under CT-guidance. Cystic fluid contents were aspirated and sporadically sent for cytologic and/or microbiologic analyses. 1–3 cm3 of contrast material (Omnipaque 180, Amersham Health, Princeton, NJ) was then instilled to outline the cavity and determine any continuity with adjacent structures
3. Results Of the seven patients in our study, one was lost to follow-up after his doxycycline treatment session. Of the six remaining patients who had at least one follow-up visit, two have undergone complete resolution of their LM (Figs. 1 and 4) without further incident (992 days mean follow-up) following an average of 1.5 treatments; the four remaining patients developed recurrent swelling. For three of the four patients with recurrent swelling, the recurrences coincided with episodes of infection. One of these three patients experienced swelling 4 years after sclerotherapy, which resolved without further incident following conservative treatment with oral antibiotics (now at 1477 days follow-up). Another of these three achieved near complete resolution, but he continues to demonstrate mild swelling of the right neck. He rarely develops infections in this area that result in increased swelling, but these resolve quickly with oral antibiotics. Because of the rarity of infection and the subtle nature of the swelling present at baseline, the family has opted to pursue conservative management only (now at 1511 days of follow-up). The last of the three patients – with swelling that was related to infection – required hospitalization for treatment with intravenous antibiotics. Although the swelling has since improved, the family is considering a second treatment with sclerotherapy (at 74 days follow-up). The patient who developed a recurrence of a fluid collection without an infectious cause (Patient #1 in Table 2) was available for only one follow-up visit at 32 days. This child is a part of the foster care system which has impaired our ability to regularly follow up with her. With regard to complications, one of our patients presented to the emergency room after developing a cellulitis which required admission to the hospital for intravenous antibiotics. He responded well and was soon discharged on an oral antibiotic course. A different patient complained of pain at the time of procedure; he was one of only two patients who did not receive general anesthesia. Notably, after appropriate intra-operative pain control, his post-operative pain was unremarkable. None of the patients reported complications such as fevers, chills, or post-operative pain.
Table 2 Clinical presentations, outcomes, and follow-up. Age listed is age at the time of initial treatment. Age/sex
Lesion size and location
Stage
Number of treatments
Peri-procedural antibiotics
Clinical results
Follow-up (days)
1 2 3
3y/F 10y/M 2m/M
II II IV
1 1 1
None Amoxicillin/clavulanate Clindamycin
Recurrence Unknown Complete resolution
32 0 1115
4
7y/F
Right neck, level II: 8 cm 5 cm 6 cm Right parotid: 4 cm 3 cm 3.5 cm Right submandibular space: 3.9 cm 3.4 cm 3 cm Left parotid and submandibular space: 4 cm 4.5 cm 4 cm
II
1
Cefazolin
1477
5
1m/F
Right mandible: 6 cm 2 cm 5 cm
IV
1
Ampicillin/sulbactam
6 7
2y/M 3y/M
Left parotid tail: 4 cm 6 cm 2.5 cm Right neck: 7 cm 6 cm 4 cm
II II
2 1
Cephalexin None
Developed swelling with infection, now with complete resolution following conservative treatment Recurrence with infection, plan for 2nd sclerotherapy treatment Complete resolution Near complete resolution, develops mild infection rarely that responds to oral antibiotics
74 868 1511
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
1129
Fig. 3. Image capture during repeat sclerotherapy procedure showing persistent but smaller lesion.
In summary, 50% of our patients with at least one follow-up visit had complete resolution of their lymphatic malformation. When the complete and near-complete resolution patients were combined, our rate of successful treatment was 67% (at 1243 days mean follow-up). In other words, 33% of our patients did not experience significant clinical improvement of their LM or developed a recurrence at the endpoint of the study (follow-up time for these patients was 53 days).
4. Discussion
Fig. 1. Photograph depicting a typical pre-treatment clinical presentation.
Fig. 2. Image capture from pictured patient during initial CT-guided sclerotherapy procedure.
Although doxycycline has been used as a sclerosant for years, its exact mechanism of action in this capacity remains unclear. It appears to induce an inflammatory reaction within the endothelial-lined cavity that leads to deposition of fibrin and collagen with eventual involution [14,15]. It also acts as an angiogenesis inhibitor by interfering with cell proliferation and migration. It does so by inhibiting matrix metalloproteinase (MMP) and suppressing vascular endothelial growth factor (VEGF)-induced angiogenesis and lymphangiogenesis [14]. Its efficacy has been previously established, and our rates of significant LM resolution are comparable with the rates of 70–100% found in the literature [7,10,13,14]. Interestingly, those patients in our series with complete resolution have the greatest length of follow-up, suggesting that use of doxycycline sclerotherapy correlates with improved results in the long-term. However, our results indicate that not all macrocystic LM respond to doxycycline sclerotherapy as one-third of our patients did not demonstrate significant clinical improvement. Complications noted in other studies, such as post-treatment hemorrhage, fever, scarring, and development of Horner’s syndrome [7,10,14,15] did not occur in our patients, although one child in our series was hospitalized for an infection requiring intravenous antibiotics. Although post-treatment systemic levels of doxycycline are elevated at supra-therapeutic levels in patients who receive a standard sclerotherapy dose (median level = 16 mg/ ml, compared to a normal therapeutic range of 1.5–2.1 mg/ml), no drug-induced side effects are noted in these patients [14]. The most significant adverse effect associated with tetracyclines is the irreversible staining of permanent teeth and bone. Doxycycline, however, has less affinity for bone and its association with dental staining is insignificant [7].
1130
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
Fig. 4. Post-treatment photograph of previously pictured patient.
Doxycycline has been reported to cause severe discomfort on injection, requiring general anesthesia in most patients [14]. This is consistent with our experience. One of the two procedures performed using local anesthesia with sedation was interrupted by patient complaints of significant pain, requiring several doses of narcotic pain medication. Post-procedure, however, the patient’s pain was easily controlled with over-the-counter analgesics. As noted in Table 1, multiple sclerosants are currently employed to treat lymphatic malformations. Bleomycin is one such drug that initially gained popularity as a sclerosing agent due to its known efficacy in the treatment of pleural effusions. Preliminary results were promising, with no major complications. One series, however, despite LM resolution rates of about 67%, led to a treatment-related death [7,16]. Its use further demands caution due to the well-established risk of pulmonary toxicity at doses greater than 500 mg [11]. OK-432 is a biologic preparation of lyophilized powder containing Streptococcus pyogenes Su strain cells (group A, type 3) treated with benzylpenicillin G that was popularized in Japan [7,11,17]. Like doxycycline, its mechanism of action is also unclear, although it appears to alter the white blood cell population, thereby inducing production of various cytokines. One such cytokine, tumor necrosis factor, increases endothelial permeability, thus allowing for egress of lymphatic fluid with resultant cell death [7]. In the United States, use of OK-432 is restricted to research protocols. Studies have shown it to be effective in treatment of macrocystic LM, although its efficacy appears limited in both microcystic lesions as well as post-surgical ones [7,18]. Reactions to treatment with OK-432 include fever, anorexia,
malaise, nausea, and hypersensitivity. Its use is contraindicated in patients with a penicillin allergy. Ethanol is a potent and widely used sclerosing agent. Its mechanism of action is via rapid cellular dehydration with precipitation of protoplasm as it comes into contact with the endothelial lining of lymphatic malformations [7]. However, it yields inconsistent and unreliable treatment outcomes, and its use is limited – particularly in small children with large lesions – due to the risk of systemic toxicity [2,14]. At doses of 1 ml/kg, patients may have increased serum ethanol levels that put them at risk for respiratory depression, cardiac arrhythmias, rhabdomyolysis, hypoglycemia, and seizures [7]. Furthermore, it cannot be injected near major nerve trunks due to its neurolytic effects, which further limits its use in the head and neck region [7,14]. Hu et al. recently found that ethanol sclerotherapy for facial venous malformations predisposes the zygomatic and temporal branches of the facial nerve to injury [19]. Sodium tetradecyl sulfate (STS), a detergent, is one of the most common sclerosing agents used; it is commercially available for this purpose. It is reported to cause less swelling than ethanol and is uniquely used for injection of orbital LM. In addition to a possible association with increased rates of infection, at least one episode of vision loss has been reported [7]. A recent case series by Shiels et al. reported on the efficacy of multi-drug percutaneous sclerotherapy. After individual aspiration of cysts, macrocystic LM were pre-treated for 2 min with 3% STS, followed by a 15 min treatment with 98% ethanol. Microcysts were also individually aspirated and then injected with doxycycline. Remarkably, 93% of 28 macrocyst patients required only one treatment session to attain complete resolution of their LM, with 100% resolution of all individually treated cysts (including both macrocysts and microcysts) within 3 days of completing treatment. The authors hypothesize that their excellent result in patients undergoing fewer than average treatments, specifically in macrocysts, is the direct result of combination therapy. They postulate that STS functions as a detergent to release transmembrane lipoproteins and increase LM membrane permeability, allowing greater penetration of ethanol, the latter of which causes intracellular protein denaturation, leading to cell death. Perhaps the most significant finding in this study was the successful treatment of microcystic lesions, albeit only those cysts large enough to be individually cannulated, aspirated, and injected with doxycycline [2]. However, microcystic and mixed LM continue to be challenging problems. Cahill et al. recently reviewed their experience with doxycycline sclerotherapy for LM in neonates and infants. For the ten patients with macrocystic LM, doxycycline at a concentration of 10 mg/ml was catheter instilled while for the seven patients with mixed LM, doxycycline at a concentration of 10 mg/ml or STS foam was directly injected into the microcyst. The majority of the mixed LM patients (6/7) were treated with a direct injection of doxycycline as opposed to a direct injection with STS foam. Their data demonstrated two-thirds of patients with macrocystic LM had significant improvement, in which a large subset of this group had near total resolution. However, less than 50% of patients with mixed LM treated with doxycycline injection noted significant improvement [13]. Our study certainly has weaknesses: it is a retrospective review with low power. Nevertheless, our results, in combination with a review of the literature, demonstrate moderate efficacy for doxycycline as a single sclerotherapeutic modality for pediatric macrocystic head and neck LM as well as the possibility of increased benefit with carefully chosen combination sclerotherapy in mixed/microcystic LM. Another potential weakness is the variable concentration of doxycycline in our study (ranging between 10 mg/ml and 20 mg/
N. Jamal et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 1127–1131
ml). Although different concentrations were used based on the volume of the macrocyst being treated, the total dose of doxycycline never exceeded 233 mg per treatment session. In addition, both ends of the range are within the limits of acceptable treatment protocols. For example, Shiels et al. used a concentration of 20 mg/ml in their study while Cahill et al. and Burrows et al. used a concentration of 10 mg/ml in their studies [2,13,14]. The relative ease and efficacy of treatment combined with the relatively low rate of treatment-related morbidity are strong points that support the superiority of percutaneous sclerotherapy as an alternative to traditional surgical extirpation. However, the sclerotherapy associated side effects cannot be entirely dismissed, as more research is shedding light onto which complications occur most frequently and which patient groups are at higher risk (i.e. infants and neonates) [13]. Interestingly, our data suggest that the rate of resolution is directly proportional to the length of followup, implying that the full therapeutic process may have not yet occurred in our patients with incomplete resolution; indeed, these patients are also the individuals with the shortest length of posttreatment follow-up. What remains to be determined is the appropriate selection, combination, and dosing regimens of specific sclerotherapeutic agents, and the specific time course in which resolution should be expected to occur for both macrocystic and microcystic/mixed LM. Conflict of interest statement The authors report no conflicts of interest in this study. References [1] R.J. Smith, Lymphatic malformations, Lymphat. Res. Biol. 2 (1) (2004) 25–31. [2] W.E. Shiels, D.R. Kang, J.W. Murakami, M.J. Hogan, G.J. Wiet, Percutaneous treatment of lymphatic malformations, Otolaryngol. Head Neck Surg. 141 (2) (2009) 219–224. [3] T.L. Kennedy, M. Whitaker, P. Pellitteri, W.E. Wood, Cystic hygroma/lymphangioma: a rational approach to management, Laryngoscope 111 (2001) 1929– 1937.
1131
[4] D. Nehra, L. Jacobson, P. Barnes, B. Mallory, C.T. Albanese, K.G. Sylvester, Doxycycline sclerotherapy as primary treatment of head and neck lymphatic malformations in children, J. Pediatr. Surg. 43 (3) (2008) 451–460. [5] C.M. Giguere, N.M. Bauman, Y. Sato, D.K. Burke, J.H. Greinwald, S. Pransky, et al., Treatment of lymphangiomas with OK-432 (Picibanil) sclerotherapy: a prospective multi-institutional trial, Arch. Otolaryngol. Head Neck Surg. 128 (10) (2002) 1137–1144. [6] M.C. Smith, M.B. Zimmerman, D.K. Burke, N.M. Bauman, Y. Sato, R.J. Smith, Efficacy and safety of OK-432 immunotherapy of lymphatic malformations, Laryngoscope 119 (1) (2009) 107–115. [7] A.I. Alomari, V.E. Karian, D.J. Lord, H.M. Padua, P.E. Burrows, Percutaneous sclerotherapy for lymphatic malformations: a retrospective analysis of patientevaluated improvement, J. Vasc. Interv. Radiol. 17 (2006) 1639–1648. [8] J.L. Acevedo, R.K. Shah, S.E. Brietzke, Nonsurgical therapies for lymphangiomas: a systematic review, Otolaryngol. Head Neck Surg. 138 (April (4)) (2008) 418– 424. [9] L.M. de Serres, K.C. Sie, M.A. Richardson, Lymphatic malformations of the head and neck. A proposal for staging, Arch. Otolaryngol. Head Neck Surg. 121 (5) (1995) 577–582. [10] H.I. Molitch, E.C. Unger, C.L. Witte, E. vanSonnenberg, Percutaneous sclerotherapy of lymphangiomas, Radiology 194 (2) (1995) 343–347. [11] D.C. Bloom, J.A. Perkins, S.C. Manning, Management of lymphatic malformations, Curr. Opin. Otolaryngol. Head Neck Surg. 12 (2004) 500–504. [12] S. Wiegand, B. Eivazi, A.P. Zimmermann, A.M. Sesterhenn, J.A. Werner, Sclerotherapy of lymphangiomas of the head and neck, Head Neck 33 (November (11)) (2011) 1649–1655. [13] A.M. Cahill, E. Nijs, D. Ballah, D. Rabinowitz, L. Thompson, N. Rintoul, et al., Percutaneous sclerotherapy in neonatal and infant head and neck lymphatic malformations: a single center experience, J. Pediatr. Surg. 46 (November (11)) (2011) 2083–2095. [14] P.E. Burrows, R.K. Mitri, A. Alomari, H.M. Padua, D.J. Lord, M.B. Sylvia, et al., Percutaneous sclerotherapy of lymphatic malformations with doxycycline, Lymphat. Res. Biol. 6 (3–4) (2008) 209–216. [15] B.M. Cordes, F.G. Seidel, M. Sulek, C.M. Giannoni, E.M. Friedman, Doxycycline sclerotherapy as the primary treatment for head and neck lymphatic malformations, Otolaryngol. Head Neck Surg. 137 (6) (2007) 962–964. [16] K.H. Kim, M.W. Sung, J.L. Roh, M.H. Han, Sclerotherapy for congenital lesions in the head and neck, Otolaryngol. Head Neck Surg. 131 (2004) 307–316. [17] W. Mol, H. Furukawa, S. Sasaki, U. Tomaru, T. Hayashi, A. Saito, Evaluation of the sclerotherapeutic efficacy of ethanol, polidocanol, and OK-432 using an in vitro model, Dermatol. Surg. 33 (December (12)) (2007) 1452–1459. [18] D.K. Gupta, M. Bajpai, M. Srinivas, A. Lall, S. Dave, S. Ogita, Efficacy of OK-432 in pediatric lymphangiomas: preliminary experience, J. Indian Assoc. Pediatr. Surg. 4 (1999) 192–195. [19] X. Hu, D. Chen, C. Jiang, Y. Jin, H. Chen, G. Ma, et al., Retrospective analysis of facial paralysis caused by ethanol sclerotherapy for facial venous malformation, Head Neck 33 (November (11)) (2011) 1616–1621.