Rheumatoid Arthritis, Baker’s Cyst, and “Thrombophlebitis” By ARMINE. GOOD From a small series of patients hospitalized with rheumatoid arthritis, six cases are presented in which Baker’s cysts were associated with pain a n d swelling of the calf, initially erroneously diagnosed as thrombophlebitis. The pathogenesis of these calf symptoms is reviewed. T Ex u n pauco extense serie de casos d e patientes hospitalisate con arthritis rheumatoidee, sex es presentate in le quales cystes d e Baker esseva associate CQll dolores e tumescentia del sura, initialmente diagnosticate i n error como thrombophlebitis. Le pathogenese d e iste symptomas sural es revistate. HE APPEARANCE of pain, swelling, and tenderness in the calf, uni- lateral ankle edema, and a positive Homans’ sign calls for a presumptive diagnosis of venous thrombosis. Although arthritis at the knee and ankle may produce these signs and symptoms, there is usually no cause for confusion when swelling and tenderness are predominant in and around the joints. In patients with arthritis, however, swelling, erythema, tenderness and pain in the calf may appear, clearly separable from articular involvement. Our recent experience indicates that these symptoms are not uncommonly manifestations of a Baker’s cyst rather than of phlebitis. In several instances, the cysts were elicited only after calf swelling and erythema subsided, or patients were properly positioned, standing with the knee in full extension. MATERIAL AND METHODS As an outcome of our experience in Case 1 with the combination of rheumatoid arthritia, a Baker’s cyst, and suspected thrombophlebitis, 45 consecutive cases, beginning with Case 1, of definite or classical rheumatoid arthritis hospitalized at Ann Arbor Veterans Administration Hospital for various reasons during the period from November 1961 through December 1962 were screened for thrombophlebitis and Baker’s cysts (table 1.). Twelve patients had popliteal cysts, either demonstrable or well documented in previous records. Eight cases had had, at one time or another, a clinical diagnosis of thrombophlebitis in the lower leg; six of these were patients with Baker’s cysts, whose histories are presented as follows: CASE1 ( AAVAH, A-8584) The patient, W. W. D., a 45-year-old male with definite rheumatoid arthritis was admitted in November 1961, five months after the onset of polyarthritis. On admission numerous joints were involved including the left knee, which was the site of effusion, increased heat, and tenderness. There was no evidence of a Baker’s cyst. On the 14th hospital day, pain, increased heat, swelling, and non-pitting edema with a positive Homans’ sign appeared in the left calf. A diagnosis of thrombophlebitis was made and the patient was treated by elevation of the leg and hot soaks. After four days, pain, heat From the Vaterans Admhktratwn Hospital, Ann Arbor, Mich., and the Depcrrtment of Medicine, University of Michigan Medical School, Ann Arbm, Mich. 56 A R T H H I ~ IASN D RIIEUhfATISM, VOI.. 7 , NO. 1 (FEBRUARY), 1964 57 R.A., BAKER’S CYST AND THROMBOPHLEBITIS 12 History of systemic corticosteroid therapy 8 (67%) History of diagnosis of thrombophlebitis of leg veins 6 (50%) 24 9 16 ( 6 7 % ) 4 (44%) Total number of episodes diagnosed as thrombophlebitis 1 1 During corticosteroid therapy 0 0 In absence of corticosteroid therapy 1 1 and edema subsided in the calf and a cystic, medial popliteal mass was delineated, measuring 4 cm. in width and extending downward 7 cm. from the transverse skin crease. This was painless, persisted for three months, decreasing slowly in size, paralleling complete disappearance of objective joint signs at the left knee. Follow-up to January 1963 has disclosed no signs or sequellae attributable to venous disease. CASE2 (AAVAH, A-663) The patient, C. E. D., a 54-year-old male with classical rheumatoid arthritis symptomatic since 1954 was admitted in 1960 because of a suspected venous thrombosis in the left calf. For five months, pain and swelling had been especially severe at both knees. Two months previously he had noted sudden painful enlargement of the left calf. The left calf was enlarged 2.5 cm. in circumference over the right and moderately tender. There was 2 f pitting edema in the left lower leg to, the mid tibia, I f similarly on the right. The left Homans’ sign was positive; no erythema or heat were present. Examination of the knees showed synovial effusion and increased heat bilaterally. With bedrest and elastic bandage wraps for a week, the calf swelling and tenderness disappeared. Numerous recorded joint examinations previously and during this admission did not disclose a popliteal cyst. The patient was readmitted for evaluation on 10-19-61 at which time there was 1+ pitting edema of the ankles but no asymmetry of the calves or other signs suggestive of venous disease. Baker’s cysts, 6 x 4 cm. on the left and 4% x 3 cm. on the right were palpated. The patient stated that the cysts had been present since the time of his discharge from the hospital in 1960. These have persisted without associated symptoms to the time of the most recent examination in November 1962 (see figs. 1 and 2). At bath knees, chronic pain, effusion, and increased skin temperature have continued. CASE3 (AAVAH A-184) The patient, F. J., a 52-year-old male with classical rheumatoid arthritis symptomatic since 1945 was seen in December of 1962 because of a flare of right knee pain and suspected recurrent phlebitis. In 1959, two months after the initial appearance of arthritis in the right knee, he was hospitalized at the Ann Arbor Veterans Administration Hospital with a diagnosis of acute thrombophlebitis, posterior compart- 5s ARMIN E. GOOD Fig. 1.-(C. E. D.), 1962-Posterior view of knees, with skin markings delineating t h e bilateral popliteal cystic masses. ment, right leg. The right calf was swollen, warm, and tender; the Honians’ sign was positive; pitting edema was present at the ankles bilaterally; and synovial effusion and tenderness were evident a t the right knee. Oral anticoagulants were started. Calf swelling cleared within several days and joint pain and swelling subsided somewhat. During admission a large, right popliteal cyst was discovered. Because of involvement of the cervical spine, the patient was readmitted to the hospital in 1980.There was no evidence of phlebitis, and a Baker’s cyst was not found. A Baker’s cyst was not detected in the course of numerous examinations between 1980 and 1982. In 1962 he experienced a recurrence of right knee pain and upper calf tenderness and swelling, reminiscent to him of the symptoms preceeding the diagnosis of phlebitis in 1959. Examination two days after onset disclosed a soft, tender, 7 x 5 cm., cystic, right, medial popliteal mass. Tenderness, increased heat, and effusion were present at the right knee. The symptoms improved following a brief period of bed rest. CASE4 (AAVAH A-23970) The patient, A. H., a 59-year-old male with classical rheumatoid arthritis symptomatic since 1937 was admitted for the first time in January 1956 because of swelling and increased heat in the left calf, where tenderness and a positive Homans’ sign were noted. The left calf appeared 30 per cent enlarged over the right. Synovial effusion and increased heat were present at both knees. He gave a history of a mass in the left popliteal space diagnosed by his doctor in 1954 as a Baker’s cyst, disappearing after needle aspiration. The patient was treated for snspected venous tliroinbosis by bedrest, elevation of the leg and oral anticoagulants. The calf signs disappeared in two weeks except for a persistent, non-tender, firm mass in the upper medial calf. This was surgically explored and found to be a thickly encapsulated .3 x 2 cm. inass high in the medial gastrocnemius containing a hematoma. Histologically this was found to be a fibrous capsule in which R.A., BAKEICS CYST AND THROMBOPHLEBITIS Fig. 2.-(C. E. D.), 1962-Left protrusion of the popliteal cyst. 59 leg externally rotated 45", showing the medial hemosiderin granules were plentifnl. A layer of fibrin was present on the inner surface and the lumen was virtually filled with an organizing hematoma. In January 1958 he was readmitted because of one week of tenderness, swelling, redness and heat in the left calf. A positive Homans' sign and 3 cm. increase in circumference of the left calf over the right were noted. The calf signs disappeared in 10 days with oral anticoagulation and bedrest. He was admitted for re-evalution in February 1962 at which time an asymptomatic left Baker's cyst, 2 x 3 cm. in size was elicited. On return visit since then the cyst has fluctuated in size, paralleling more or less the degree of pain and effnsion of the left knee, occasionally disappearing completely. CASE5 ( AAVAH A-2343) The patient, 0. P., a 61-year-old white male with classical rheumatoid arthritis symptomatic since 1938 was admitted in December 1961 with a diagnosis of thrombophlebitis of the left calf. Three weeks previously severe pain and swelling had appeared at the left knee and calf, followed in several days by a tender, bluish discoloration and swelling of the medial left calf. On admission the left knee was irritable and difficult to move because of pain. An edematons, warm, tender swelling was noted in the left calf exhibiting yellow-green discoloration, 8 x 16 cm. in size, extending downward from the left popliteal space. A well healed, linear surgical scar was noted in the medial le't calf. Synovial effnsion was marked at the left knee and minimal on the right. The calf pain and induration cleared slowly within one month, coincidentally with improvem m t in the signs at the left knee joint. Records were then obtained from an outside hospital where a diagnosis of thrombophlebitis was made in December 1957 became of left calf tenderness and a positive Homans' sign. Active arthritis was present in both knees at the time. Within two weeks the calf signs disappeared except for a sausage shaped mass approximately X! cm. in ARMIN E. GOOD 60 length extending from the popliteal space into the calf. This was explored and found to be a mesothelially lined cyst with a densely collagenous wall. Injection of dye into the knee joint demonstrated communication with the cyst. A diagnosis of Baker’s cyst was made. Follow-up to 1962 has disclosed no reappearance of a discrete, fluctuant mass in the left calf or popliteal area. However, in the light of the well documented history of a popliteal cyst simulating thromhophlehitis, the abrupt calf swelling and pain followed by hemorrhagic discoloration in 1961 suggested a recurrent popliteal cystic protrusion, perhaps rupturing spontaneously. The admitting diagnosis of thromhophlebitis was abandoned. CASE6 (AAVAH A-8397) The patient, J. P. P., a 74-year-old white male was admitted in November 1961 with classical rheumatoid arthritis symptomatic since 1953. Objective changes were present in numerous joints, especially the knees, where warmth, tenderness and synovial effusion were noted. Erythema and ill defined swelling was present over the medial aspect of the left calf, where dilated superficial varicosities were seen distally. The Homans’ sign was negative. There was bilateral pitting ankle edema, most marked in the left leg. A diagnosis of probable phlebothronibosis of the left calf was made. Four days later a cystic mass was delineated in the medial gastrocnemius extending down 7 cm. from the transverse skin crease (fig. 3 ) . This was reaclily reducible by manual compression, producing a bruit audible on auscultation with a stethoscope over the patella. It was felt that the previous diagnosis of phlehothromhosis was an error. Over the subsequent year, the cyst decreased somewhat in size and produced no significant symptoms. DISCUSSION Cystic tumors of the popliteal space have been variously called popliteal cysts, synovial cysts, posterior hernias of the knee joints, Baker’s cysts, semimembranous bursae or gastrocnemius semimembranous bursae. Characteristically these present as cystic swellings distal to the transverse skin crease, closely associated with the medial insertion of the gastrocnemius (See Fig.’s 1-4). Tenderness is usually minimal and symptoms may be entirely absent. At times the mass may be palpable only with the patient erect and the knee in full extension. The cysts are occasionally lax1 and difficult to elicit proximally because of the overlying popliteal fat pad and deep fascia. They may present as a mass in the lower calf as distal as the Achilles tend or^.^-^ In the calf, the cystic nature may be obscured by overlying r n ~ s c l e . ~ Wilson, Eyre-Brook, and Francis demonstrated a patent communication between the knee joint and the gastrocnemius-semimembranous bursae in 17 of 30 cadaver dissecti0ns.l In the presence of arthritis of the knee, the escape of joint fluid into the bursa via this communication is generally regarded as the pathogenesis of popliteal ~ y s t s . Adams ~ - ~ in 18407 a d Baker in 188S5 emphasized the etiologic importance of rheumatoid arthritis and other causes of joint effusion in the production of popliteal cysts. Moreover, in the first case described in his original report, Baker was concerned with the differential diagnosis between a popliteal cyst and a venous thrombosis, when he encountered a patient presenting with unilateral ankle edema, calf swelling, and dilatation and tortuousity of the superficial veins distal to the knee. Others have associated a popliteal cyst with edema of the lower leg and foot.”.” The popliteal vein, medial distally, crosses over the RA., BAKER’S CYST AND THROMBOPHLEH~TJS 61 Fig. &-(J. P. P.), 1961-Medial aspect of left lower leg showing Baker’s cyst presenting as sausage shaped swelling arising from popliteal space. At the lower pole, saphenous varicosities are visible. artery in the popliteal space, subject to compression between a popliteal cyst and the artery. Thus, venous compression by inapparent popliteal cysts may account for certain instances of unexplained dependent leg edema, a problem encountered in a proportion of patients with rheumatoid arthritis.1° In general Baker’s cysts found in association with rheumatoid arthritis have been remarkably asymptomatic, tending to wax and wane in concert with the synovial effusion of the knee. The pathogenesis of the exceptional attacks of calf pain and diffuse swelling associated with popliteal cysts remains somewhat conjectural. Firstly, is the possibility that these attacks are due to venous thrombosis, facilitated by Baker’s cysts. This has not, however, been our experience in cases 4 and 5, where operative exploration disclosed no thrombosis, nor that of Harvey and C o r c o ~who , ~ described four patients operated upon because of large popliteal cysts extending to the mid calf and below. They noted that these cysts were readily confused with thrombophlebitis of the leg veins, the initial diagnosis in each instance. At surgical exploration, they 69 ARhlIN E. GOOD Fig. 4.-Posterior view of knees, from patient with rheumatoid arthritis, showing, on the right leg, the typical medial location of a popliteal cyst, distal to the transverse skin crease. The cystic m a s s produced 110 symptoms and the patient was unaware of its presence. did not feel that venous disease contributed significantly to the symptoms, although in one patient a single small thrombosed vein was discovered. Apart from venous compression, other possible explanations for calf symptoms include rupture of the cyst, hemorrhage into the cyst, inflammation or synovitis of the cyst wall, and even associated rheumatoid granulomata. Others have commented upon the rapid appearance and enlargement of cysts, suggesting rupture and escape of free fluid.4 Haggart found the walls of popliteal cysts to be densc near the knee but generally friable and easily ruptured distally.s Rupture is also suggested by the fact that a large proportion of specimens show no definite lining of synovial cells, but merely an intermingling of lymphocytes, plasma cells and h i~ tio c y te s "and ~ ~ polymorphonuclear cells.3 Histiologically, there is often evidence of hemorrhage, such as clumps of intracellular hemosiderinl and fibrin deposition." Within the cyst wall, an irregular palisade-like fibroblastic formation resembling that seen in a rheumatoid nodule has been o b s e r ~ e d . ~ , ~ In our material, episodes of calf pain followed periods of increased symptoms of arthritis at the knees. Synovial effusion was a prominent finding in each instance. In case 1 in which the knee was observed closely before and after the appearance of the mass, the pain and inflammatory signs may have been due to acute enlargement or dissection of the cyst. In cases 4 and 5, there is evidence for hemorrhage, although in the latter case this may have been induced by anticoagulants. In retrospect a t least, it appears that the diagnosis of thrombophlebitis was R.A., BAKER’S CYST AND THHOMBOPHLEBITIS 63 erroneous in each of our six cases. Follow-up has been notable for the absence of sequelae of thrombophlebitis such as palpable thrombosed veins, pulmonary embolism, and signs of venous insufficiency. Fever and tachycardia were absent in each case. It is noteworthy that none of our six patients manifested any of the follcwing, more or less definitive, signs of thrombophlebitis: tenderness along the course of a deep vein; visible, tender, erythematous, or palpable cords; cyanosis of the calf; dilatation of collateral veins; edema above the knee; and arterial spasm. The diagnoses of thrombophlebitis, in each of our cases, were made on the basis of several of the following presumptive signs or symptoms: pain in the calf, diffuse calf tenderness, Homans’ sign, increase of skin temperature, erythema, and edema distal to the knee. These signs have demonstrated merit in the early diagnosis of venous thrombosis, especially in certain high-risk situations, such as in the postoperative hypercoagulable state. However, in dealing with a patient with synovial effusion or recent flare of synovitis in the knee, these presumptive signs of thrombophlebitis should be interpreted with due skepticism. In order to prevent prolonged hospitalization due to needless anticoagulant therapy, the wise course in such a setting may be to treat the synovitis aggressively and the possible phlebitis conservatively. Thus intra-articular injection of compound F can be combined with bed rest, elevation of the leg, elastic wrappings and close observation. When the calf signs subside completely in 24 to 48 hours, it may be inferred that the diagnosis of deep venous thrombosis is in error. As edema subsides, the delineation of a Baker’s cyst may provide an alternative explanation for calf symptoms. Finally, our experience with this consecutive, unselected series of patients with rheumatoid arthritis does not support the suggestion4 that corticosteroid therapy increases the incidence of Baker’s cysts. Of 24 patients with a history of knee involvem,ent but without popliteal cysts, 16 (67 per cent) had received systemic corticosteroids. Of the 12 patients with Baker’s cysts, all of whom had had objective evidence for arthritis of the corresponding knee, eight ( 67 per cent) had received corticosteroids. Likewise the acute episodes of calf symptoms, originally diagnosed as thrombophlebitis, in the overall group of 45 patients did not appear related to the use of corticosteroids. Of 10 such episodes among eight patients, only four occurred during corticosteroid therapy (table 1). SUMMARYAND CONCLUSIONS 1. A survey of 45 consecutively hospitalized patients with rheumatoid arthritis disclosed a history of a diagnosis of thrombophlebitis of the leg veins in 12 and of Baker’s cysts in eight. 2. Six of the patients who had both “thrombophlebitis” and Baker’s cysts are presented, illustrating the thesis that the presumptive signs of thrombophlebitis may be mimicked by the sudden enlargement of, hemorrhage into, or perhaps rupture of the cyst. 3. In each case, synovial effusion of the corresponding knee was obvious; 64 ARMIN E. GOOD an exacerbation of arthritis in the knee usually preceded the episodic calf symptoms. 4. Although retrospectively the diagnosis of thrombophlebitis per se was probably erroneous in each of these cases, Baker’s cysts may occasionally interfere with the venous return in the leg. 5. Our data do not suggest that popliteal cysts are precipitated or make worse by prolonged corticosteroid therapy. REFERENCES 1. Wilson, P. D., Eyre-Brook, A. L., and Francis, J. D.: A clinical and anatomical study of the semimembranosus bursa in relation to popliteal cyst. J. Bone & Joint Surg. 20:963, 1938. 2. Kersley, G. D., Barber, H. S., Cregan, J. C . F., and Gibson, H. J.: Degenerative rheumatoid changes. J. Bone & Joint Surg. 36B:238, 1954. 3. Burleson, R. J., Bickel, W. H., and Dahlin, D. C.: Popliteal cyst: A clinicopathological survey. J. Bone & Joint Surg. 38A: 1265, 1956. 4. Harvey, J. P., Jr., and Corcos, J.: Large cysts in lower leg originating in the knee occurring in patients with rheumatoid arthritis. Arth. & Rheumat. 3:218, 1960. 5. Baker, W. M.: The formation of abnormal synovial cysts in connection with the joints. St. Bartholomew’s Hosp. Rep. 21:177, 1885. 6. Meyerding, H. W., and VanDemark, R. E.: Posterior hernia of the knee (Baker’s cyst, popliteal cyst, semimembranosus bursitis, medial gastrocnemius bursitis and popliteal bursitis ) . J.A.M.A. 122:858, 1943. 7. Adams: Chronic rheumatic arthritis of the knee joint. Dublin J . M. Sc. 17: 520, 1840. 8. Haggart, G. F.: Posterior hernia of the knee joint: a cause of internal derangement of the knee. J. Bone & Joint Surg. 20:363, 1938. 9. Lewin, P.: The knee and related structures. Philadelphia, Lea & Febiger, 1952, p. 271. 10. Short, C. L., Bauer, W., and Reynolds, W. E.: Rheumatoid Arthritis. Cambridge, Mass., Harvard University Press, 1957, pp. 323-5. Armin E. Good, M.D., Chief, Rheumtology Section, Veterans Administration Hospital, Ann Arbor, Michigan; Assistant Professor of Internal Medicine, University of Michigan Medical Center, Ann Arbor, Michigan.