Annals of the Rheumatic Diseases, 1988; 47, 423-427

Case report

Subacute bacterial endocarditis in a patient withankylosing spondylitisRICHARD A HOPPMANN,' CHRISTOPHER M WISE,'VENKATA R CHALLA,2 AND JAMES E PEACOCK1

From the Departments of 'Medicine and 2Pathology, Bowman Gray School of Medicine of Wake ForestUniversity

SUMMARY A 57 year old white man with aortic insufficiency and previously undiagnosedankylosing spondylitis, who developed subacute bacterial endocarditis (SBE), is described.Emergency aortic valve replacement was necessary, and the aortic valve pathology showeddiffusely fibrosed and thickened valve leaflets with bacterial vegetations. This is the first recordedcase of SBE in ankylosing spondylitis.

Mallory is generally credited with first reporting theassociation between ankylosing spondylitis and aor-tic insufficiency in 1936.' Graham and Smythe in1958 reported on 519 patients with ankylosinespondylitis, 10% of whom had aortic insufficiency.Numerous other reports have followed recordingthe association of ankylosing spondylitis and aorticinsufficiency, with prevalences of aortic insuf-ficiency up to Graham and Smythe's orginal figureof 10%. The aortic insufficiency of ankylosingspondylitis results from fibrous shortening andthickening of the aortic valve leaflets and dilatationof the aortic root. Extension of the fibrosis into theventricular septum and the conduction system givesrise to the various forms of heart block seen inankylosing spondylitis.3

In the light of the prevalence of associatedvalvular abnormalities in patients with ankylosingspondylitis one would expect SBE to be occasionallyencountered, but an extensive review of publishedwork has failed to show any well recorded cases ofSBE complicating ankylosing spondylitis. Thereexist only four incomplete cases, in which neitherthe diagnosis of ankylosing spondylitis nor SBE issubstantiated.46 The following case represents thefirst documented case of SBE in ankylosing spondy-litis.

Accepted for publication 26 October 1987.Correspondence to Dr Richard A Hoppmann. RheumatologySection, East Carolina University School of Medicine, Greenville,NC 27858-4354, USA.

Case report

A 57 year old white man presented to his localemergency room on 2 February 1986 with acuteonset of left eye blindness and an erythrocytesedimentation rate of 60 mm/h. He was given 100mg of prednisone and transferred to our faculty.There was no history of prior visual changes,headaches, stiffness, or jaw claudication. He deniedfever, weight loss, myalgias, or arthritis, except foroccasional low back pain.

His past medical history included peptic ulcerdisease, prostatitis, mild aortic insufficiency, andplacement of an atrioventricular sequential pace-maker for syncopal episodes secondary to Wenc-kebach arrhythmia in December 1981.

Physical examination was remarkable for aninfected left conjunctiva. Fundoscopic examinationof the left eye showed a pale retina, and the eye wasblind. Temporal arteries were non-tender. Heartexamination showed a 2/6 systolic murmur and a 2/6diastolic murmur, both heard best in the aortic regionand apparently unchanged from previous evalua-tions. There were no significant skin lesions.Laboratory investigation showed haemoglobin 150g/l, white blood cell count 10*7x 109/l, and erythro-cyte sedimentation rate 40 mm/h.He was admitted to the hospital and given 60 mg

of prednisone a day. A right temporal artery biopsywas normal. Doppler studies of the carotid arteriesshowed minimum plaque formation without signifi-

423

copyright. on N

ovember 19, 2021 by guest. P

rotected byhttp://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.47.5.423 on 1 May 1988. D

ownloaded from

424 Hoppmann, Wise, Challa, Peacock

cant stenotic flow. Two dimensional echocardiogra-phy showed thickening of the aortic valve withoutevidence of vegetations. A rheumatoid factor (latexand sheep cell agglutination test (SCAT)) andantinuclear antibody (ANA) tests were negative.Prednisone was discontinued, and the patient wasdischarged with a diagnosis of left retinal arteryocclusion, probably secondary to an embolic event.He was readmitted on 11 March 1986 with a two

day history of intermittent left shoulder and armpain associated with diaphoresis. Physical examina-tion was unchanged from the previous stay inhospital. Electrocardiogram showed totally pacedrhythm and chest x ray revealed mild cardiomegalyand increased vascular markings. Laboratory studieswere as follows: haemoglobin 115 g/l, white bloodcell count 11-6x 109/l normal urine analysis, normalblood chemistries, erythrocyte sedimentation rate79 mm/h, and creatine phosphokinase 91 U.

Because of an increase in the creatine phospho-kinase to 231 U with a 15% MB band the patientunderwent coronary arteriography, which showedno significant coronary artery disease, 4+ aorticinsufficiency, and mild aortic valve calcification.Upper gastrointestinal series, barium enema, andsigmoidoscopy were normal. Serum iron and totaliron binding capacity were consistent with cnronicdisease. Four sets of blood cultures were initiallynegative. Serum protein electrophoresis did notshow a monoclonal protein. Left temporal arterybiospy was normal.

Further interrogation revealed back pain that wasworse with inactivity. Schober test was positive fordecreased lumbar flexion. Sacroiliac joint filmsshowed fusion of the sacroiliac joints and syn-desmophyte formation (Fig. 1). HLA-B27 waspositive. He was given naproxen with some im-provement and discharged with a diagnosis of

Fig. 1 Anteroposteriorradiograph showing fusion of thesacroiliac joints bilaterallv andSYndesmophyte formation.

copyright. on N

ovember 19, 2021 by guest. P

rotected byhttp://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.47.5.423 on 1 May 1988. D

ownloaded from

Subacute bacterial endocarditis and ankylosing spondylitis 425

ankylosing spondylitis but was readmitted threedays later with left shoulder and neck pain,myalgias, fever to 38-9°C, and dyspnoea on exer-tion. Haemoglobin was 91 g/l, white blood cell count8-2x 109/l, and the erythrocyte sedimentation rate 72mm/h. It was noted that one blood culture drawn onthe day of his previous discharge was now growingStaphylococcus epidermidis. Repeat two dimen-sional echocardiogram showed a large vegetation onthe aortic valve. Rheumatoid factor was now posi-tive with a latex test titre of 1/640 and a SCAT of1/128. An ANA test was positive at 1/100 with ahomogeneous pattern.

He was given intravenous vancomycin and gen-tamycin for presumed infectious endocarditis. Aftertwo weeks of treatment he experienced acutecardiac decompensation secondary to progressiveaortic insufficiency necessitating emergency aorticvalve replacement. Culture of valve tissue wasnegative.

Pathological examination of the aortic valveshowed fibrosed and thickened leaflets with vegeta-tions containing clusters of Gram positive cocci (Figs2 and 3). Antibiotics were continued for a furtherthree weeks. He was discharged and two weeks laterhad resumed his normal daily activities.

AORTAIntimaMediaAdventitia

L. Atrium Fig. 2 On the left is a lowpowerphotomicrograph ofone oftheaortic leaflets and on the right adiagrammatic illustration ofthechanges in the aortic valve inankylosingspondylitis (modifiedfrom Bulkley B HandRoberts W C?). The diagram showsthefibrous thickening in the aorticadventitia extending into themitral ring and into the aorticleaflet. It has been modified toinclude the site ofthe vegetation(V). The part of the diagrambetween the arrows represents theactual tissue specimen at left. Notethe two hyalinised nodules (*)along the valve leaflet and the

r-ti distally located vegetation (V) in-i-/. /the tissue specimen corresponding

to the same structures in theAnterior diagram.

copyright. on N

ovember 19, 2021 by guest. P

rotected byhttp://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.47.5.423 on 1 May 1988. D

ownloaded from

426 Hoppmann, Wise, Challa, Peacock

Fig. 3 High power photomicrograph ofthe vegetationshows numerous colonies of very large cocci. The bloatedappearance ofthe cocci is attributed to antibiotic treatment.(Gram stain).

Discussion

The four previously reported cases of ankylosingspondylitis complicated by SBE lack clear docu-mentation of either diagnosis, and none includesaortic valve pathology. In 1956 Blumberg andRagan, while reporting on the natural historyof 'rheumatoid spondylitis', briefly mentioned apatient who developed SBE.4 In addition to aorticinsufficiency, this patient also had aortic stenosisand mitral stenosis, making rheumatic heart diseasea more likely diagnosis as neither of these lattervalvular lesions are recognised as a complication ofankylosing spondylitis. In 1975 Cosh and colleaguesmentioned a case of SBE from an unreferencedseries of patients with ankylosing spondylitis, inwhich one patient died from renal failure after beingtreated for bacterial endocarditis.5 Valkenborgh et

al, in 1976, presented 25 patients with various formsof inflammatory arthritis and heart disease.6 Two ofthese may have represented SBE in patients withankylosing spondylitis, but they were not reportedin sufficient detail to be certain of the diagnosis.Thus the present case represents the firstdocumented case of SBE in a patient with ankylos-ing spondylitis. The underlying aortic valve pathol-ogy with fibrous thickening throughout the leafletand fibrous nodule formation is characteristic of thatreported in ankylosing spondylitis.3There are several explanations why SBE has not

been reported in association with ankylosing spon-dylitis. One possibility would be that SBE truly is arare complication of the valvular disease seen inankylosing spondylitis. This seems unlikely, how-ever, as the degenerative changes of the aortic valvedescribed in patients with ankylosing spondylitis(i.e., fibrous thickening with nodule formation) arewell characterised predisposing valvular lesions ofendocarditis.7 8A second explanation could be that when SBE

occurs in the setting of ankylosing spondylitis theunderlying spondylitis is not appreciated, and thusan association between endocarditis and spondylitisis not recognised. This would be consistent withBergfeldt's discovery of previously undiagnosedspondylitis in a group of patients with permanentpacemakers.9Another possibility is that the diagnosis of super-

imposed SBE may not be made in patients withknown ankylosing spondylitis. The course of aorticinsufficiency in patients with ankylosing spondylitisis generally slowly progressive and acute haemody-namic compromise requiring valve replacement israre. t1 Nevertheless, acute cardiac decompen-sation in such patients may be incorrectly attributedto progression of valvular disease or malignantarrhythmia rather than to superimposed SBE. Inaddition, the musculoskeletal complaints which areoften the first symptoms of SBE12 may be attributedto extraspinal involvement of ankylosing spondylitisrather than to SBE.

Finally, it could be that SBE is encountered inankylosing spondylitis but simply not reported.Although this would seem to be a likely explanation,SBE has not been reported in morbidity andmortality figures in longitudinal studies of ankylos-ing spondylitis.

In summary, this case report documents theoccurrence of SBE in a patient with ankylosingspondylitis. Aortic valve pathology showed bacterialvegetations on a diffusely fibrosed and thickenedvalve typical of the valvular disease in ankylosingspondylitis. This case was illustrative of severalclinical points. Firstly, both conduction abnormali-

copyright. on N

ovember 19, 2021 by guest. P

rotected byhttp://ard.bm

j.com/

Ann R

heum D

is: first published as 10.1136/ard.47.5.423 on 1 May 1988. D

ownloaded from

Subacute bacterial endocarditis and ankylosing spondylitis 427

ties and valvular disease occur in patients withpreviously undiagnosed ankylosing spondylitis.Secondly, if patients have ankylosing spondylitis andnew musculoskeletal complaints, especially in thesetting of systemic manifestations or acute valvulardecompensation, this should alert the physician tothe possibility of SBE. Thirdly, the value of serialechocardiography in diagnosing SBE was shown.Finally, valvular replacement can be successfullyperformed in the setting of SBE in ankylosingspondylitis.

References

1 Mallory T B. Case records of the Massachusetts GeneralHospital. N Engl J Med 1936; 214: 690-8.

2 Graham D C, Smythe H A. The carditis and aortitis ofankylosing spondylitis. Bull Rheum Dis 1958; 9: 171-4.

3 Bulkley B H, Roberts W C. Ankylosing spondylitis and aorticregurgitation. Description of the characteristic cardiovascularlesion from study of eight necropsy patients. Circulation 1973;18: 1014-27.

4 Blumberg B, Ragan C. The natural history of rheumatoidspondylitis. Medicine (Baltimore) 1956; 35: 1-31.

5 Cosh J A, Gerber N, Barritt D W, Jayson M I V. Cardiaclesions in Reiter's syndrome and ankylosing spondylitis. AnnRheum Dis 1975; 34: 195.

6 Valkenborgh P, Dequeker J, Gielen F, DeGeest H. Arthritis

and heart lesions: a study of 25 cases with pericarditis orvalvular lesions associated to inflammatory joint disease. ActaCardiol (Brux) 1976: 4: 269-76.

7 Kayc D. Infective endocarditis. Baltimore: University ParkPress, 1976: 6-8.

8 Scheld W M, Sande M A. Endocarditis and endovascularinfections. In: Mandell G L. Douglas Jr R G, Bennett J E. eds.Principles and practice ofinfectious diseases. 2nd ed. New York:Wiley, 1985: 504-30.

9 Bergfeldt L. HLA B27-associated rheumatic discasc withsevere cardiac bradyarrhythmias. Clinical features in 223 menwith permanent pacemakers. Am J Med 1983; 75: 21(-5.

10 Demoulin J C, Lespagnard J, Bertholet M, Soumagne D. Acutefulminant aortic regurgitation in ankylosing spondylitis. AmHeart J 1983; 105: 859-61.

11 Stewart S R, Robbins D L, Castles J J. Acute fulminant aorticand mitral insufficiency in ankylosing spondylitis. N Engl J Med1978; 299: 1448-9.

12 Churchill M A Jr, Geraci J E, Hunder G G. Musculoskeletalmanifestations of bacterial endocarditis. Ann Intern Med 1977;87: 754-9.

13 Lehtinen K. 76 patients with ankylosing spondylitis seen after30 years of disease. Scand J Rheumatol 1983; 12: 5-11.

14 Carette S, Graham D, Little H, Rubenstein J, Rosen P. Thenatural disease course of ankylosing spondylitis. ArthritisRheum 1983; 26: 186-90.

15 Hart F D, Maclagan N F. Ankylosing spondylitis: a review of184 cases. Ann Rheum Dis 1955; 14: 77-82.

16 Wilkinson M, Bywaters E G L. Clinical fcatures and course ofankylosing spondylitis: as seen in a follow-up of 222 hospitalreferrcd cases. Ann Rheum Dis 1958; 17: 209-28. copyright.

on Novem

ber 19, 2021 by guest. Protected by

http://ard.bmj.com

/A

nn Rheum

Dis: first published as 10.1136/ard.47.5.423 on 1 M

ay 1988. Dow

nloaded from