вход по аккаунту


Gonococcal arthritis. Clinical features correlated with blood synovial fluid and genitourinary cultures

код для вставкиСкачать
Official Journal of The American Rheumatism Association
Section of The Arthritis Foundation
Gonococcal Arthritis
Clinical Features Correlated with Blood, Synovial Fluid
and Genitourinary Cultures
Kenneth D. Brandt, Edgar S. Cathcart and Alan S. Cohen
The clinical picture was analyzed in 31 patients with acute arthritis and
proven gonococcal infection. Patients with positive blood cultures and
those with positive synovial fluid cultures were mutually exclusive. However, these two groups were indistinguishable on the basis of mean
duration of illness prior to initial examination, the number of joints involved or the magnitude of synovial fluid leukocytosis. Notably, features
that have been particularly associated with gonococcemia-eg, chills, a
migratory polyarticular prodrome and typical skin lesions, were present
in most cases. Thus, on clinical grounds none of the patients in the
present series could be clearly categorized into either the bacterernic
or septic joint syndromes of gonococcal arthritis.
Of the many systemic complications of
genitourinary infection with N gonorrhea,
the most common is arthritis, which arises
as a consequence of the hematogenous seeding of synovium (1-3). T w o syndromes of
gonococcal arthritis have recently been de-
From the Arthritis and Connective Tissue Disease Section, Boston University Medical Center,
Departments of Medicine, Boston City Hospital and
University Hospital and the Evans Department of
Clinical Research.
Grants in support of these investigations have
been received from the United States Public Health
Service, National Institute of Arthritis and Metabolic Diseases (AM-04599 and TI-AM-5285), the
General Clinical Research Centers Branch of the
Division of Research Resources, National Institutes
of Health (RR-533), the Massachusetts Chapter of
the Arthritis Foundation and the Arthritis Foundation.
KENNETH D BRANDT, MD: Special Research Fellow,
National Institutes of Health, and Assistant Profes-
sor of Medicine, Boston University School of Medicine: EDGAR s CATHCART, MD: Associate Professor of
Medicine, Boston University School of Medicine;
ALAN s COHEN, MD: Conrad Wesselhoeft Professor
of Medicine and Director, Medical Services, Boston
City Hospital and Thorndike Laboratories.
Presented in part at the Eighteenth Interim
Scientific Session of the American Rheumatism
Association, Pittsburgh, Pennsylvania, December 8-9,
Address reprint requests to Kenneth D. Brandt,
MD, Arthritis Section, University Hospital, 750
Harrison Avenue, Boston, Massachusetts 021 18.
Submitted for publication November 26. 1973;
accepted February 25, 1974.
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
Table 1. Results of Cultures in 31 Patients with Arthritis and Gonococcal Infection
Blood Cultures
No. of
No. of
Synovial Fluid Cultures
No. of
Genitourinary Cultures
No. of
'Cultures were obtained from two joints of 2 patients
lineated and considered to represent sequential stages in the evolution of the
disease (4).
In one, a bacteremic syndrome, the duration of illness before initial evaluation is
brief. Shaking chills, fever and skin lesions
are prominent and, although multiple joints
are involved, synovial effusions are seldom
noted. In contrast, other patients with gonococcal arthritis have been described with
a septic joint syndrome marked by a tendency to monarticular arthritis and the absence of chillls, fever and skin lesions. It
has been felt that the septic joint syndrome
develops after a relatively asymptomatic
bacteremic episode, when organisms have
incubated in synovium for a sufficient period to produce a purulent effusion (5,6).
With these clinical descriptions in mind,
we recently examined our experience with
patients who had arthritis associated with
gonococcal infection. Contrary to the reports cited above, none of the patients in
this series could be clearly separated on
clinical grounds into one or the other of
the above syndromes.
The patients included in the present study fultilled three criteria: all had a) acute a r t h r i t i 6 i e .
joint symptoms for less than two weeks prior to
initial medical evaluation; b) concurrent bacterio-
logic proof of gonococcal infection and c) prompt
response of arthritis to antibiotic therapy (defined
as a decrease of 50% or greater in the synovial fluid
leukocyte count within 48 hours and/or complete
resolution of the joint effusion within 10 days of
the onset of treatment). Thirty-one patients satisfying these criteria were seen in the past 10 years.
The patients were classified into three groups,
according to the site from which evidence of gonococcal infection was obtained (Table 1):
Group I consisted of 7 patients from whom N
gonorrhea was cultured from the blood.
Group I1 consisted of 15 patients with positive
synovial fluid cultures.
Group 111 was comprised of 9 patients in whom
the sole bacteriologic evidence of gonococcal infection was obtained from cultures of the cervix,
vagina, prostatic secretions or anus.
General Characteristics
Twenty-one of the 31 patients (68%)
were women; distribution according to sex
was comparable for all three groups. The
youngest patient in the series was 16, the
oldest 62 and the mean and median ages
were 27.3 and 23.5 years, respectively. Although males tended to be somewhat older
than females, the three groups were essentially similar with respect to age.
Six patients (2 from Group 11 and 4 from
Group 111) were pregnant; 2 were in the
first, 1 was in the second and 3 were in the
third trimester of gestation (Table 2).
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
Table 2. Comparison of Clinical Features in 31 patients with Arthritis and Gonococcal Infection
Clinical Feature
Number of patients
General characteristics
Age, mean (yrs)
Interval, onset of illness to initial
evaluation, mean (days)
Systemic reaction to infection
Fever, maximum, initial 24 hours of
hospitalization, mean (OF)
Peripheral blood white cell count, initial 24 hours
of hospitalization, mean (cells/mm*)
Patients demonstrating signs of gonococcemia
Skin lesions
Migratory polyarthritis
Bacteriologic Data
Synovial fluid cultures from a single joint
were obtained from 15 of the 15 patients
in Group 11. By definition, all yielded N
gonorrhea. In addition, two positive synovial fluid cultures were obtained from both
of the remaining patients in this group
(Table 1). Although not all patients in
Groups I and 111 were felt to have joint
effusions, synovial fluid cuItures, which
proved to be sterile, were taken from single
joints of 5 patients in Group I and from 4
in Group 111.
None of the patients with positive synovial fluid cultures had positive blood cultures (Table l), although at least two blood
cultures were obtained from every patient
in Group 11. Three or more blood cultures
were taken from 11 of the 15 patients.
Hence the complete absence of positive
blood cultures in Group I1 was probably
not due to inadequate attempts to culture
the organism. In contrast however blood
cultures were obtained from only 2 patients
in Group 111 (both of whom also had sterile synovial fluid cultures).
By definition all patients in Group 111
had active genitourinary infection with N
gonorrhea. In addition, concurrent gonorrhea was not uncommon in the other two
groups, and the gonococcus was cultured
from the genitourinary tract or anus of 5
patients from Group I and 5 from Group
I1 during the episode of acute arthritis.
Seven patients, moreover, including 3 of the
above, had been treated for bacteriologically proven gonorrhea at periods ranging
from 1 month to 11 years before the onset
of arthritis. No patient in the series however had experienced a previous episode of
gonococcal arthritis or symptoms suggestive
Interval Before Initial Evaluation
Articular pain, with or without swelling,
was the initial manifestation of illness in
all but one patient and was generally the
factor responsible for leading the patient to
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
Fig 1. Relationship of fever to site from which
positive culture of N gonorrhea was obtained.
Group I, positive blood culture; Group 11, positive
synovial fluid culture; Group 111, only genitourinary tract culture positive.
consult a physician. Although constitutional
symptoms-eg, malaise, fatigue, shaking
chills or fever-invariably appeared within
2 days after the onset of joint pain, their
presence did not necessarily hasten the
initial evaluation and in several cases even
recurrent shaking chills and drenching
sweats did not move the patient to seek
medical care.
T h e interval between the onset of illness
and initial evaluation was similarly brief
for patients in Groups I and 11, with a
mean of 2.3 and 3.4 days, respectively (Table 2). I n some cases from both groups the
onset was extremely acute. Thus, 2 patients
in Group I and 4 in Group I1 had articular
symptoms for 24 hours or less prior to admission to the hospital. Patients from Group
111, on the other hand, were ill longer before seeking medical attention (mean: 7.5
days), and none had articular symptoms for
less than 4 days before hospitalization.
Fever, Chills and Peripheral
Blood Leukocyte Count
T h e maximum oral temperature rec
orded during the initial 24-hour period of
hospitalization was noted (Table 2). I n the
occasional case in which rectal rather than
oral temperatures were recorded, 1°F was
subtracted to approximate the oral temperature. T h e mean initial temperature was
103.4OF for patients in Group I, 101.9OF
for those in Group I1 and 100.2'F for those
in Group 111 (Figure 1). Patients in Group I
also had a somewhat greater degree of peripheral blood leukocytosis on admission to
hospital (mean: 17,500 white blood cells/
cumm) than those in Groups I1 and I11
(12,200 and 11,300 white blood cells/cu mm,
respectively). Shaking chills were present in
slightly over half the patients in this series,
and no appreciable difference in their prevalence was seen when the three groups were
Skin Lesions Consistent
with Gonococcemia
Skin lesions compatible with gonococcemia (7,s) were relatively uncommon (Table 2). In 6 patients (4 in Group I, 2 in
Group 11) erythematous pustules were noted
on the initial examination or within the
following 24 hours. All pustules were sterile
on culture and Gram stains of the lesions
revealed no organisms.
Joint Disease
T h e character of the arthritis in all of
the above patients was examined. Attention
was paid to both the patient's description
of the joint disease during the interval before medical evaluation and the findings
at the time of the initial examination.
Prodrome. I n most instances joints that
were symptomatic prior to initial evaluation tended to remain so until treatment
was instituted. However, migratory polyarticular disease (defined as either pain or
swelling of joints in sequential fashion,
with improvement in some affected joints
prior to institution of treatment) occurred
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
Fig 2. Relationship of clinical features of bacteremia to site from which positive culture of N
gonorrhea was obtained. Group I, positive blood
culture; Group II, positive synovial fluid culture;
Group 111, only genitourinary tract culture positive.
The numbers represent the patients in each group
with and without clinical evidence of bacteremia
chills, cutaneous lesions consistent with
gonococcemia or migratory polyarthritis.
in 11 patients, including 1 in Group I, 7
in Group I1 and 3 in Group 111.
Despite its infrequency in Group I (Table 2), migratory polyarthritis has previously been associated with gonococcal bacteremia and has been reported to occur as
a prodromal feature of gonococcal arthritis
(5,9,10). Its relationship to other features,
generally considered to represent clinical
expressions of gonococcemia-eg, chills and
skin lesions was therefore examined in the
present series. A strong concurrence of
migratory polyarthritis, chills and skin lesions was not apparent and the concurrence
of all three features was not noted in any
of the above patients. On the other hand,
at least one component of this triad was
... ..
.. ...
Fig 3. Relationship between joint involvement
observed at time of initial evaluation and site
from which positive culture of N gonorrhea was
obtained. Group I, positive blood culture; Group
II, positive synovial fluid culture; Group Ill, only
genitourinary tract culture positive.
present in all patients with proven gonococcemia (Group I), 11 of the 15 patients
in Group I1 and 6 of the 9 in Group I11
(Figure 2).
Joint Involvement on Initial Evaluation.
Monoarticular joint disease was relatively
uncommon and occurred in only 5 patients
(2 in Group I, 2 in Group I1 and 1 in
Group 111). There was no apparent tendency toward single joint infections in patients with positive synovial fluid cultures
or for multiple joint involvement in patients with positive blood cultures (Figure 3).
When joint disease was related to other
clinical features considered to indicate the
bacteremic stage, none of the 5 patients
with monarticular arthritis had temperatures over 102OF (mean: 100.6OF) in the
initial 24-hour period of hospitalization. Although 14 of the 26 patients who had two
or more joints involved had temperatures
of 102OF or greater (mean: 103.4OF), no apparent relationship was found between
magnitude of fever and number of joints
involved on initial examination in those
with polyarthritis (Figure 4). There also
was no apparent relationship between the
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
11,000 to 100,000 cells/cu mm. T h e average
synovial fluid leukocyte count in the series
was 82,000 white blood cells/cu mm, with
no major difference between the three
.=Group 1
No relationship was found between the
synovial fluid white cell count and duration
of illness. In several instances counts of
lOO,OOO/cu mm or greater were observed in
2 ' 99
fluids from patients of Group I1 who had
articular symptoms for 24 hours or less.
Joints Involved, Initial Examination
Furthermore when patients with involveFig 4. Relationship of fever to number of joints ment of a single joint were compared with
involved on initial evaluation of patients with those having polyarticular disease, no difgonococcal arthritis. Group I, positive blood cul- ferences were apparent in the magnitude of
ture; Group 11, positive synovial fluid culture; the synovial fluid white cell count. SimilarGroup 111, only genitourinary tract culture positive
ly no clear correlation could be found befor N gonorrhea.
tween the synovial fluid leukocyte count
results of the synovial fluid cultures.
number of abnormal joints at the time of
fluids with leukocyte counts over
initial evaluation and either the prevalence
mm were sterile, while gonoof chills, skin lesions or the magnitude oE
from three fluids containthe peripheral blood leukocyte count.
cells/cu mm).
Synovial Fluid. Arthrocentesis was performed on all joints in which effusions
were considered to be present. Twenty-five Influence of Gonorrhea and of Pregnancy
synovial fluids, obtained at the time of ini- on the Clinical Picture
tial evaluation from 4 patients in Group I,
As noted above, patients in Group 111
all 15 in Group I1 and 4 in Group 111, were displayed a longer duration of illness prior
analyzed. From 2 patients in Group I1 to initial evaluation and tended to be febfluids from two joints (both of which were rile less frequently than the others, raising
positive on culture) were examined. Leuko- the question of whether genitourinary incyte counts of the paired fluids from these fection with the gonococcus might have
2 cases showed good agreement, with sam- modified the clinical picture of gonococcal
ples from the knees of 1 patient containing arthritis. However, patients in Groups I
700 and 1,400 white blood cells/cu mm, and 11 who had concurrent genitourinary
respectively, and those from the ankle and infection, or in whom a previous episode
wrist of the other patient both showing of gonorrhea had been documented, were
marked synovial inflammation, with 322,000 indistinguishable from the others with reand 120,000 white blood cells/cu mm, re- spect to the prevalence of any of the features considered. There was also no indicaspectively.
Although 3 patients in Group I were not tion that the clinical features of gonococcal
considered to have effusions when examined, arthritis in pregnancy tended to be any dif4 others had inflammatory synovial effu- ferent from those noted in the series as a
sions, with leukocyte counts ranging from whole.
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
The present series of cases of gonococcal
arthritis resembles others in the antibiotic
era (4,ll) with respect to age and sex distribution. Since the joint disease in each of
the above patients responded rapidly and
completely to appropriate antibiotic therapy, it is unlikely that any suffered from
nongonococcal postvenereal arthritis (1 0,12)
rather than, or in addition to, gonococcal
The two clinical syndromes of gonococcal arthritis described (4,6) have been considered to represent extremes in the evolution of the disease, from a bacteremic polyarticular stage to subsequent localized
mono- or oligoarticular synovial infection.
The present series is notable insofar as every patient showed features of both the
bacteremic and septic joint syndromes. Neither the interval between onset of illness
and initial evaluation nor any other feature
(Table 2) permitted accurate clinical differentiation between patients with positive
blood cultures and those with positive synovial fluid cultures. Indeed, of the three
clinical signs particularly associated with
gonococcemia-ie, chills, migratory polyarthritis, typical skin lesions, at least one appeared in most patients in this series. That
the concurrence of all three was not seen
in any patient may have been due, at least
in part, to the fact that skin lesions were
observed in only 19%. In other series skin
lesions have been noted in approximately
50% of the patients (4,6). In any event the
above data emphasize that clinical evidence
of gonococcemia may be much more common than bacteriologic evidence.
'The findings in the present series are at
variance with the concept that the septic
joint syndrome arises only after a relatively
lengthy incubation period of the gonococcus in the synovium of patients who earlier
incurred an asymptomatic bacteremic stage.
There was no indication that patients with
monarticular arthritis had responded to
bacteremia less symptomatically than did
those with polyarticular disease when the
prevalence of chills or skin lesions, the
prevalence or magnitude of fever (Figure 4),
the degree of peripheral blood leukocytosis
or the severity of other constitutional symptoms were considered. Moreover, at least as
indicated by the magnitude of the synovial
fluid leukocytosis, no relationship was observed between severity of the joint infection and duration of disease in the aspirated joint.
Even though they could not be separated
on clinical grounds, patients with positive
synovial fluid cultures in the present series
were, without exception, mutually exclusive. Notably, this bacteriologic dichotomy
is consistent with results of other series (4,6)
and appears to be characteristic of gonococcal arthritis. Thus in contrast, blood cultures were positive in 39y0 of 18 patients
with proven staphylococcal and 57Y0 of 18
patients with proven streptococcal arthritis
seen by us during the period of the present
study (13).
To date, the frequency of bacteremia in
patients with gonorrhea is not known. It
has been estimated that approximately
0.6% of infected females develop gonococcal
arthritis (14) and that 3% of females and
0.7% of males treated for gonorrhea develop skin lesions of gonococcemia (15). Since
multiple joint membranes are seeded with
bacteria during gonococcemia, it may be
asked why gonococcal arthritis is not more
common. Presumably, in most instances
host responses are sufficient to deal with the
infection locally in the synovial membrane.
Where such local responses are insufficient,
pyogenic arthritis may develop-gradually,
as in patients with the typical septic joint
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
syndrome, or more rapidly, as in several
patients in the present study. In the latter
case, where patients were observed to have
pyogenic arthritis concurrent with evidence
of gonococcemia, it may have been that the
infecting strain of gonococcus was particularly virulent or host defenses in the synovium were impaired. Such possibilities could
account for differences between the clinical
picture of gonococcal arthritis in Boston,
on the one hand, and in Cleveland (4) or
Seattle (6), on the other, where the illness
has been separated more readily into either
the bacteremic or septic joint syndrome.
Hewes HF: Two cases of gonorrheal rheumatism with specific bacterial organisms in
in the blood. Boston Med Surg J 131:515516, 1894
Keefer CS, Parker F, Myers WK: Histologic
changes of the knee joint of various infections. Arch Pathol 18:199-215, 1934
Sharp JT: Arthritis associated with genitourinary infection, Vistas in Connective Tissue Disease. Edited by JC Bennet. Springfield, Illinois, Charles C Thomas, 1968, p 155
Keiser HL, Ruben FL, Wolinsky E, et al:
Clinical forms of gonococcal arthritis. N
E n d T Med 279:234-240. 1968
5. Calabro JJ: Gonococcal arthritis in the
young. N Engl J Med 279:1002, 1968
6. Holmes KK, Counts GW, Beaty HN: Disseminated gonococcal infection. Ann Intern
Med 74:979-993, 1971
7. Abu-Nassar H, Hill N, Fred HL, et al:
Cutaneous manifestations of gonococcemia:
review of 14 cases. Arch Intern Med 112:
731-737, 1963
8. Ackerman AB, Miller RC, Shapiro L:
Gonococcemia and its cutaneous manifestations. Arch Dermatol 91:227-232, 1965
9. Wheeler JK, Heffron WA, Williams RC, Jr:
Migratory arthralgias and cutaneous lesions
as confusing initial manifestations of gonorrhea. Am J Med Sci 260:150-159, 1970
10. Wright V: Arthritis associated with venereal
disease: A comparative study of gonococcal
arthritis and Reiter’s syndrome. Ann Rheum
Dis 22:77-90, 1963
11. Grabar WJ, Sanford JP, Ziff M: Sex incidence of gonococcal arthritis. Arthritis
Rheum 3:309-313, 1960
12. Partain JO, Cathcart ES, Cohen AS: Arthritis associated with gonorrhea. Ann Rheum
Dis 27:156-162, 1968
13. Goldenberg D, Brandt K, Cohen AS: Unpublished observations
14. Holmes KK, Wiesner PJ, Pedersen AHB:
T h e gonococcal arthritis-dermatitis syndrome. Ann Intern Med 75:470-471, 1971
15. Barr J, Danielsson D: Septic gonococcal dermatitis. Br Med 1 1:482-485. 1971
Arthritis and Rheumatism, Vol. 17, No. 5 (September-October 1974)
Без категории
Размер файла
498 Кб
features, clinical, correlates, genitourinary, arthritis, culture, synovial, gonococcal, fluid, blood
Пожаловаться на содержимое документа