Prevertebral Tendinitis: How to Avoid Unnecessary Surgical Interventions

Transcription

1 The Laryngoscope VC 2012 The American Laryngological, Rhinological and Otological Society, Inc. Prevertebral Tendinitis: How to Avoid Unnecessary Surgical Interventions Georg Philipp Hammer, MD; Robert Vollmann, MD; Peter Valentin Tomazic, MD; Josef Simbrunner, DI, MD; Gerhard Friedrich, MD Objectives/Hypothesis: Prevertebral tendinitis is an inflammatory process that affects the cervicothoracic prevertebral muscles. Because of its clinical presentation and imaging features in computed tomography scans, prevertebral tendinitis can easily be mistaken for deep cervical abscess formation. Totally different therapy regimens require clinical and diagnostic pathways for sufficient differentiation between those two pathologic entities. Study Design: Case series with comparison. Methods: In 10 patients with prevertebral tendinitis, we evaluated the symptoms, laboratory reports, and radiological imaging findings. We compared these data to 65 patients with a deep cervical abscess formation. The basic radiologic imaging procedure was contrast-enhanced computed tomography. For detection of prevertebral tendinitis, we performed magnetic resonance imaging with diffusion-weighted images and calculated the apparent diffusion coefficient map. Results: Patients with prevertebral tendinitis complained of severe neck pain, globus sensation, and neck stiffness. Diffusion-weighted images showed a typical benign prevertebral effusion. Computed tomography scans showed amorphous calcifications in the tendon of the prevertebral muscles. The C-reactive protein values were slightly increased in patients with prevertebral tendinitis, and white blood cell count remained normal. In comparison to patients with deep cervical abscess formation, the C-reactive protein and white blood cell count was significantly lower (P <.05) in the prevertebral tendinitis cases. Conclusions: Prevertebral tendinitis should be considered when patients suffer from neck pain, neck stiffness, and globus sensations despite low signs of inflammation in the laboratory report. To confirm the diagnosis, the best imaging feature is magnetic resonance imaging with diffusion-weighted images and apparent diffusion coefficient map. Key Words: Head and neck, prevertebral tendinitis, retropharyngeal abscess, benign prevertebral effusion, magnetic resonance imaging, computed tomography. Level of Evidence: 4. Laryngoscope, 122: , 2012 From the ENT University Hospital Graz (G.P.H., P.V.T., G.F.) and Department of Radiology (R.V., J.S.), Medical University of Graz, Graz, Austria. Editor s Note: This Manuscript was accepted for publication March 12, The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Georg Philipp Hammer, MD, ENT University Hospital, Medical University of Graz, Auenbruggerplatz 26, 8036 Graz, Austria. georg.hammer@medunigraz.at DOI: /lary INTRODUCTION Acute neck pain, globus sensation, swallowing problems, and a stiff neck are the symptoms that are usually found in patients with a deep cervical abscess formation. Typically, the pharynx shows a protrusion of the pharyngeal wall with signs of inflammation of the mucosa. Treatment guidelines advise surgical drainage in addition to intravenous application of antibiotics and nonsteroidal anti-inflammatory drugs (NSAIDs). 1,2 Acute calcific tendinitis of the longus colli muscle is a rare, self-limiting, inflammatory disease, which was first described by Hartley 3 in 1964, redefined by Weinberg and Scott 4 in 1982, and is currently known under a number of synonyms, such as prevertebral or retropharyngeal tendinitis The underlying pathophysiological reason is a calcium hydroxyapatite deposition in the longus colli tendon assumed to induce acute inflammation of the longus colli muscle tendon insertion. 3 Because of the anatomic localization of inflammatory changes, we prefer the name prevertebral tendinitis (PT). Due to similar symptoms, clinical presentation, and imaging features it is quite easy to misdiagnose this rare disorder for a deep cervical abscess formation near skull base. 21 Since the first reports of PT, 3,4 radiological imaging procedures have changed fundamentally; some authors proposed magnetic resonance imaging (MRI) or contrast enhanced computed tomography (CT) for detection of PT. 5,17 21 In their recent literature review, Aydil et al. 22 mentioned a large number of patients with noninfectious neck and craniofacial pain disorders who were treated inappropriately with antibiotics or by surgical intervention. The aim of our study was to call attention to this rare disease and to offer diagnostic guidelines to prevent unnecessary surgical intervention in the future. MATERIALS AND METHODS In the period from June 2009 to November 2011, six male and four female adult patients (>18 years of age) with PT were

2 TABLE I. Overall Results Comparing Demographics, Clinical Symptoms, and Laboratory Reports Between the Prevertebral Tendinitis Group and the Abscess Group. Group Tendinitis (n ¼ 10) Abscess (n ¼ 65) Age, yr 46.9 (615.8) 49.1 (618.3) Sex, male/female 6/4 34/31 Fever, no. (%) 3 (30) 22 (33) Deep neck pain, no. (%) 10 (100) 64 (99) Globus sensation, no. (%) 6 (60) 45 (69) Neck stiffness, no. (%) 7 (70)* 1 (2)* Dyspnea, no. (%) 1 (10) 6 (9) Trismus, no. (%) 1 (10) 12 (18) WBCC, median (range), g/l 10.2 ( )* 13.2 ( )* CRP, median (range), mg/l 38.5 ( )* 99.0 ( )* *Significant results (P <.05). WBCC ¼ white blood cell count; CRP ¼ C-reactive protein. included in this study. None of these patients suffered from a chronic inflammatory or malignant disease. Seven patients received MRI, and in six out of seven MRI examinations, additional diffusion-weighted imaging (DWI) was performed and the apparent diffusion coefficient (ADC) map was calculated. The other three patients received only contrast-enhanced computed tomography (CT) because of contraindication to MRI. As published elsewhere, 5 MRI examinations were performed with a 1.5-Tesla system (Siemens Magnetom Espree; Siemens, Erlangen, Germany). The imaging protocol included a diffusion-weighted single-shot spin echo-echo-planar sequence acquired in the anterior-posterior commissure (diffusion gradient b values of 500 and 1,000 s/mm 2, repetition time [TR] 5,000 ms, echo time [TE] 114 ms, slice thickness 6 mm with no gap, matrix of pixels, and field of view of 230 mm); fluidattenuated inversion recovery (TR/TE 9,770/99 ms, inversion time 2,200 ms); and T2-weighted turbo spin-echo sequences (TR/TE 4,500/85 ms). For diffusion-weighted MRI, the diffusion gradients were subsequently and separately applied in three orthogonal directions, for a total acquisition time of 97 seconds. Trace images were then generated and ADC maps calculated with a congruent software tool (Syngo; Siemens). CT studies (collimation: mm; pitch: 1.05; section thickness: 5 mm; increment: 3 mm) from the maxilla to the aortic arch were performed with a 64-slice CT scanner (Somatom Sensation 64; Siemens) before and after administration of intravenous contrast agent (50 ml iopromide; Bayer HealthCare, Leverkusen, Germany) by a power injector (Ulrich Medical, Ulm, Germany) at a rate of 3 ml/s and a scanning delay of 40 seconds after the start of the injection. A saline flush of 50 ml was admitted after the contrast agent s injection at the same rate. The laboratory reports obtained on the day of the CT or MRI examinations were reviewed for C-reactive protein (CRP) and white blood cell count (WBCC). We chose these parameters because they are included in routine laboratory blood samples and were available for each patient. At our institution, the normal value of CRP is <8 mg/l and the normal value of WBCC is <11.3 g/l. Finally, we retrospectively analyzed the laboratory reports of 65 adult patients with a deep cervical abscess formation who were treated in our clinic from 2005 to 2011, and compared those results with the PT patients laboratory reports. For statistical analysis, we used SPSS statistical software version 19.0 (IBM, Armonk, NY). We performed the Mann- Whitney U test, as the data were not distributed normally. Statistical significance was defined as a P value of <.05. This study was approved by the institutional review board of the Medical University of Graz. Written informed consent for neck CT or MRI with contrast agent administration was obtained from all patients before the procedure. RESULTS The age of our 10 patients with PT ranged from 28 to 77 years (mean age, 46.9 years; standard deviation [SD] ). All patients suffered from severe neck pain, and 70% had a stiff neck. We also detected swallowing problems (60%), and in some cases fever, dyspnea, and trismus (Table I). MRI in all patients with PT revealed a prevertebral effusion (Fig. 1). Eight out of 10 patients showed amorphous calcifications (Fig. 2) in the tendon of the longus colli muscle. In one case, calcifications could not be identified at all because of a very strong beam hardening artifact in CT caused by a dental prosthesis. Furthermore, all patients showed an asymmetric swelling of the longus colli muscle. Only in the six patients who received DWI was it possible to assess the quality of the prevertebral fluid accumulation and to detect the benign retropharyngeal effusion, which is typical for PT (Figs. 3 and 4). In four cases, we performed a follow-up MRI after successful antiphlogistic therapy with NSAIDs over 6 days. Here the prevertebral swelling could no longer be detected, the effusion vanished, and the patients did not suffer from any complaints. Compared to the nonpathologic values, CRP was slightly increased in all patients with PT, ranging from 0.6 to 86.6 mg/l (median, 38.6 mg/l; mean, 42.8 mg/l; SD ). The WBCC, however, remained normal in Fig. 1. PT-Case 8: hyperintense swelling of the prevertebral space (arrow) is revealed by T2-weighted magnetic resonance imaging sequences (sagittal view). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] 1571

3 Fig. 4. Apparent diffusion coefficient (ADC) map at the same level as Figure 3: high signal in the prevertebral space (arrow). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] Fig. 2. PT-Case 10: characteristic calcifications at the tendon of the longus colli muscle (arrow) are shown by unenhanced computed tomography of the neck at the level of the second vertebral body. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] most cases, ranging from 5.6 to 13.2 g/l (median 10.2 g/ L; mean 9.7 g/l, SD 6 2.8). In 65 patients with deep cervical abscess formation (mean age, 49.1 years; SD ), the CRP value ranged from 3.0 to mg/l (median, 99.0 mg/l; mean, mg/l; SD ), the WBCC ranged from 5.4 to 33.5 g/l (median, 13.2 g/l; mean, 14.1 g/l, SD 6 5.7). The difference in CRP and WBCC was statistically significant (P <.05) between the PT group and the abscess group (Figs. 5 and 6, Table I). According to the described symptoms, we found an interesting difference between the two groups: two out of 65 patients (3%) suffered from a stiff neck, whereas 70% of the PT patients reported this symptom. Because of the assumption of abscess formation, five patients with PT were initially treated with intravenous antibiotics (cephalosporins), and two of them were surgically drained as previously described. The cephalosporins were given on average for 7 days, but in one case only for 3 days because of the corrected diagnosis after additional MRI. The other five patients recovered without any antibiotic treatment and only with intravenous application of NSAIDs. In two out of three patients who received only CT examinations, the prevertebral inflammatory changes were misdiagnosed as an abscess, and transoral surgical drainage was performed. Here, no pus protruded in the course of intervention, and for that reason no swabs were taken for culture. The diagnosis Fig. 3. Diffusion-weighted imaging of the neck (case 2): high signal in the prevertebral space (arrow). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] Fig. 5. C-reactive protein (CRP) (mg/l): comparison between deep cervical abscess group (n ¼ 65) and prevertebral tendinitis group (n ¼ 10). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] 1572

4 Fig. 6. White blood cell count (cells/ll): comparison between deep cervical abscess group (n ¼ 65) and prevertebral tendinitis group (n ¼ 10). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] was changed accordingly after surgery by pathologic report (histology: fluid accumulation consisting of serous and fibrinous components). DISCUSSION When assuming a deep cervical abscess formation near the skull base, it is mandatory to be aware of PT as a differential diagnosis and what information is needed to detect this rare, noninfectious, musculoskeletal pain disorder. The totally different therapy regimens for both pathologic entities, however, make a clear and expeditious diagnostic pathway indispensable, especially with regard to unnecessary surgical intervention. Because of its paucity and the obvious difficulty of correct diagnostic pathways, the literature about PT is poor and contains predominantly case reports 6 12 ; a prospective study is still missing. Since Hartley s 3 and Weinberg and Scott s 4 articles in the 1960s and 1980s, the possibilities in radiologic imaging have changed fundamentally and opened different diagnostic pathways. 5,13,15,17 21 The musculus longus colli, longus capitis, rectus capitis anterior, and rectus capitis lateralis are the anterior vertebral muscles of the neck. These muscles make up the bulk of the prevertebral space, which is delimited anteriorly by the deep layer of the deep cervical fascia. Acute PT is described as a benign, self-limited, inflammatory condition of those prevertebral muscles. Mostly, the superior tendon fibers of the longus colli muscle are affected. 3,4,12 Homma et al. 14 demonstrated a foreign body inflammatory response to deposits of hydroxyapatite crystals in their evaluation of the tissue specimen taken from prevertebral space, with routine and polarized light microscopy, scanning electron microscopy, and energy-dispersive spectrometry. The typical clinical features are acute or subacute onset of pain and stiffness in the neck with odynophagia. After antiphlogistic therapy alone and physiotherapeutic interventions, the clinical symptoms vanish within a few days. 11,16 The age of our 10 PT patients ranged from 28 to 77 years, which is in accordance with other authors who postulate an accumulation of this disease in the third through sixth decade of life. 7,15 That was the reason why we compared our PT patients only with adult patients with a deep neck infection, although cervical abscess formations are also common in young children and adolescents. Referring to the underlying pathophysiologic mechanisms and the average onset age of PT, a chronic physiological load of the cervical spine could be the main reason for generating this noninfectious cervical pain disorder. As described in former case reports, our patients blood samples showed a normal count or a slight elevation of WBCC as well as slightly elevated CRP. 5,11,14,16 When treating a patient for an assumed retropharyngeal abscess, this unusually low elevation of inflammatory signs, in comparison to typical cervical abscess formations, should be considered. In two cases, surgical drainage of the supposed inflammatory formation was performed without any purulent discharge; only fluid accumulation consisting of serous and fibrinous components was documented in the histologic report. The classic imaging findings are soft tissue swelling of the prevertebral muscles and amorphous calcifications anterior to the body of the first and second cervical vertebra. The best diagnostic value for these lesions is given by CT. In MRI, calcifications appear as subtle hypointense structures in gradient echo-weighted sequences, which are very unspecific. The dystrophic calcifications in this condition are thought to have metabolic components. Additional risk factors, such as repetitive trauma, recent injury, ischemia, inflammation, or tissue necrosis, appear to play a role. However, the degree of calcification is very variable. Another important imaging finding is prevertebral effusion. In our cases, this prevertebral fluid accumulation was found in every case. The importance of this finding is the possibility of mistaking this effusion for an abscess formation, which may be an indication for surgery. In three cases, we could offer MRI examination with DWI and ADC mapping (Figs. 3 and 4). In those sequences, the fluid reveals a high signal, which is typical for serous fluid and makes an abscess unlikely. DWI is based on microscopic movement of water molecules and is defined as b value, increasing with diffusion weight. Under normal conditions, water molecules move freely in the extracellular space. The movement is restricted in abscess formations caused by necrotic material, cell debris, and inflammatory cells leading to the high viscosity of pus. These changes lead to restricted water diffusion with an increased signal on DWI and low ADC values. A serous fluid accumulation, on the other hand, leads to increased ADC values, as in our cases. The signal of serous fluid in DWI depends on the b values. For example, b values of about 500 s/mm 2 reveal a high signal caused by a T2 shine- through effect (Fig. 3). We also recommend the term benign prevertebral effusion, to distinguish this fluid accumulation from abscess formations. 5,17,

5 As described previously, the best way to assess the quality of the prevertebral effusion is MRI. In CT scans or plain radiographs, it could sometimes be quite difficult to make a clear diagnosis, especially when calcifications are discrete or cannot be detected by beam hardening artifacts. Sometimes it is also helpful to look for an asymmetric swelling of the longus colli muscle, as we detected in all of our cases, which does not typically appear in patients with cervical abscess formation. Depending on the clinical manifestation and course, we recommend a follow-up examination after antiphlogistic therapy within 5 to 7 days, especially when MRI is not available. In our clinical routine, we established a diagnostic algorithm to differentiate as fast as possible between these two pathologic entities if the above mentioned symptoms are present: after a precise anamnesis (e.g., former trauma of the cervical spine), patients usually get a CT scan. In combination with the described unspectacular inflammation signs in the laboratory reports and the clinical presentation (severe neck pain and stiffness), the diagnosis of a PT can be made, although some inexperienced radiologists may assume a retropharyngeal abscess formation. Therefore, we start intravenous antiphlogistic therapy, but we do not perform surgical intervention for drainage. Antibiotic therapy is given in primarily unclear cases, especially when inflammatory signs in the laboratory report are decisive. Unless the radiologist can make a diagnosis from the CT scan, we perform an MRI with DWI and ADC mapping mostly within the first 2 days after the first contact with the patient to confirm the diagnosis. Of course, our study has some limitations. The rarity of PT makes a statistically sufficient analysis of an adequate number of patients difficult. Additionally, we also had to learn from our own mistakes, especially after the inadequate attempt of surgical treatment. In their review, Aydil et al. 22 mention PT as a self-limited, lessknown, noninfectious disease causing severe craniofacial pain sensations. According to the available literature and to our own experience, PT needs antiphlogistic therapy with NSAIDs and a few days of clinical surveillance, including physical protection without any empiric antibiotic treatment or even inappropriate surgical drainage. For all these reasons, we advocate the importance of a precise knowledge of the diagnostic pathway in cases of suspected PT. CONCLUSION Because of its clinical presentation and imaging features in CT scans, PT can easily be mistaken for an abscess formation. However, the totally different therapy regimens of both pathologic entities make a clear and expeditious diagnostic pathway indispensable, especially regarding unnecessary surgical intervention. According to the experience with our patients, the differential diagnosis of PT should be considered, particularly when inflammation signs in the laboratory report are unspectacular and the patient complains of severe neck pain, globus sensation, and particularly neck stiffness. To reveal the benign PT and confirm the diagnosis of PT, the best imaging feature is MRI with DWI and ADC mapping. In cases where MRI is unavailable, we recommend CT scans to detect typical prevertebral calcifications. Additionally, we want to emphasize the importance of intense cooperation between the radiologist and the ear, nose, and throat specialist in every case. Acknowledgments The authors want to thank Drs. M. Starmuehler and H. P. Sallegger, who both made a valuable contribution to this study when performing their master s thesis. BIBLIOGRAPHY 1. Schuler PJ, Cohnen M, Greve J, et al. Surgical management of retropharyngeal abscesses. Acta Otolaryngol 2009;129: Brook I. Current management of upper respiratory tract and head and neck infections. Eur Arch Otorhinolaryngol 2009;266: Hartley J. Acute cervical pain associated with retropharyngeal calcium deposit. A case report. J Bone Joint Surg Am 1964;46: Weinberg S, Scott RA. Retropharyngeal tendinitis. Laryngoscope 1982;92: Vollmann R, Hammer G, Simbrunner J. Pathways in the diagnosis of prevertebral tendinitis. Eur J Radiol 2012;81: Kupferman TA, Rice CH, Gage-White L. Acute prevertebral calcific tendinitis: a nonsurgical cause of prevertebral fluid collection. Ear Nose Throat J 2007;86: Kaplan MJ, Eavey RD. Calcific tendinitis of the longus colli muscle. Ann Otol Rhino Laryngol 1984;93: Shin DE, Ahn CS, Choi JP. The acute calcific prevertebral tendinitis: report of two cases. Asian Spine J 2010;4: Bladt O, Vanhoenacker R, Bevernage C, Van Orshoven M, Van Hoe L, D Haenens P. Acute calcific prevertebral tendinitis. JBR-BTR 2008;91: DeMaeseneer M, Vreugde S, Laureys S, Sartoris DJ, DeRidder F, Osteaux M. Calcific tendinitis of the longus colli muscle. Head Neck 1997;19: Khurana B. Calcific tendinitis mimicking acute prevertebral abscess. J Emerg Med 2012;42: Park SY, Jin W, Lee SH, Park JS, Yang DD, Ryu KN. Acute retropharyngeal tendinitis: a case report with unusual location of calcification. Skeletal Radiol 2010;39: Hall FM, Docken WP, Curtis HW. Calcific tendinitis of the longus colli: diagnosis by CT. AJR Am J Roentgenol 1986;147: Homma Y, Itoi A, Muta T, et al. Acute posterior head pain in adult: a case series. Cases J 2009;23: Gelenicke J, Salomonsen M, Hiviid C. Retropharyngeal tendinitis: radiographic and magnetic resonance imaging findings. Acta Radiol 2006;47: Ring D, Vaccaro AR, Scuderi G, Patheria MN, Garfin SR. Acute calcific retropharyngeal tendinitis. Clinical presentation and pathological characterization. J Bone Joint Surg Am 1994;76: Vollmann R, Bohlsen D, Simbrunner J. Benign prevertebral edema as expression of retropharyngeal tendinitis MRI image of a rare entity [in German]. Rofo 2010;182: Ellika SK, Payne SC, Patel SC, Jain R. Acute calcific tendinitis of the longus colli: an imaging diagnosis. Dentomaxillofac Radiol 2008;37: Eastwood JD, Hudgins PA, Malone D. Retropharyngeal effusion in acute prevertebral tendinitis: diagnosis with CT and MRI imaging. AJNR Am J Neuroradiol 1998;19: Offiah CE, Hall E. Acute calcific tendinitis of the longus colli muscle: spectrum of CT appearance and anatomic correlation. Br J Radiol 2009;82: Kito S, Morimoto Y, Tanaka T, et al. Utility of diffusion weighted images using fast asymmetric spin-echo sequences for detection of abscess formation in the head and neck region. OralSurg Oral Med Pathol Oral Radiol Endod 2006;101: Aydil U, Kizil Y, Koybasioglu A. Less known non-infectious and neuromusculoskeletal system-originated anterolateral neck and craniofacial pain disorders. Eur Arch Otorhinolaryngol 2012;269: