Materials and methods

Between January 1999 and July 2002, we identified prospectively 40 consecutive patients with subclinical valvitis. During this period, 189 attacks of acute rheumatic fever were observed; 171 being initial attacks and 18 recurrences. The diagnosis was made according to the modified Jones criterions by considering history, physical examination, chest X-ray and electrocardiographic findings, white blood cell count, erythrocytic sedimentation rate, C-reactive protein, antistreptolysin-O titer and throat cultures.⁶

Echocardiographic examination of all patients was undertaken at the time of the diagnosis, using a General Electric Vingmed System Five Performance echocardiographic scanner. We consider silent carditis to be present when Doppler echocardiographic interrogation reveals valvar insufficiency in the absence of clinical evidence of carditis. Doppler echocardiographic guidelines for diagnosis of pathological mitral and aortic insufficiency have previously been established.^{7, 8} Five criterions were defined for diagnosis of pathologic mitral regurgitation:

• Length of the colour jet greater than 1 cm.
• The colour jet identified in at least two planes.
• When seen posterolaterally, the colour jet having a mosaic appearance.
• Persistence of the jet throughout systole.
• Peak velocity of mitral regurgitation exceeding 2.5 mm/s is considered pathologic.

Pathologic aortic regurgitation was considered present when:

• A substantial colour jet seen in two planes extending well beyond the valvar leaflets.
• Holodiastolic high velocity spectral envelope.
• Peak velocity of aortic regurgitation exceeding 2.5 mm/s.

The murmurs are deemed innocent when well-localized, low-grade, vibratory or musical, and occupying a brief period of early systole.

All patients with arthritis or chorea had had an initial attack of rheumatic fever, and all with arthritic findings were treated with anti-inflammatory therapy, except three patients who presented after the acute phase with a normal sedimentation rate. These patients had been treated with acetylsalicylic acid before admission, so were included amongst those receiving acetylsalicylic acid for subsequent analysis (see below). No anti-inflammatory treatment had been given to patients with chorea. The patients could be divided into three groups according to the treatment they received. Thus, we had patients treated with acetylsalicylic acid, those receiving steroids, and the patients with chorea, who had received no anti-inflammatory treatment. All patients were followed at two-week intervals by assessment of the erythrocytic sedimentation rate and physical examination during the acute attack, and monthly thereafter. The mean period of follow-up was 18.1 ± 13.9 months, with a range from 1 to 41 months.

Statistical analysis

Data were analyzed using SPSS 8.0 computer software. Results are expressed as mean values plus or minus the standard deviation. The period of follow-up, and the time for resolution of the valvar regurgitation among the groups of patients receiving different forms of treatment, were compared using the Kruskal-Wallis test. Categorical data in the three groups were compared by the use of the χ² test. The level of significance was set at 0.05 (2-sided).

Results

The ages of the patients ranged from 7 to 16 years, with a mean of 11.1 ± 2.5 years. Of the patients, 18 were male (45%), and 22 were female (55%). The major findings within the Jones criterions were arthritis in 29 patients, chorea in 10 patients, and arthritis and erythema marginatum in one patient. In 21 patients, we heard well-localized, low-grade, vibratory musical, and brief early systolic murmurs, which consequently were deemed to be innocent. All patients had normal chest X-rays. In 5 patients (16.6%), the electrocardiogram revealed first-degree heart block. The erythrocytic sedimentation rate ranged from 3 to 140 mm/h, with a mean 54.5 ± 34.4 mm/h. The titres of anti-streptolysin-O ranged from 180 to 3170 units/l with a mean of 704 ± 679 units/l (Table 1).

Table 1. Baseline characteristics and laboratory findings of our patients.

Left ventricular dimensions as measured by M-mode echocardiography were corrected to body surface area, as were left ventricular systolic functions and left atrial dimensions. Of the patients, 33 had mitral regurgitation, 6 patients had combined mitral and aortic regurgitation, and one patient had aortic regurgitation (Table 1). The colour jet of the mitral regurgitant flow was posteriorly directed in all patients. None of our patients had valvar thickening, leaflet prolapse, or sub-valvar morphological changes.

Treatment

The patients with chorea were not treated with anti-inflammatory drugs such as acetylsalicylic acid or steroids. In total, 17 patients were treated only with acetylsalicylic acid. An additional 13 patients were treated with prednisone, with or without acetylsalicylic acid. If deemed necessary, prednisone, at a dose of 2 mg/kg, was given over a period of only 10 days. Acetylsalicylic acid, at a dose of 80–100 mg/kg, was given for a period of 3 to 4 weeks, either in isolation or after treatment with prednisone.

Prognosis

The mean period of follow-up was 18.1 ± 13.9 months, with a range from 1 to 41 months. In patients treated with prednisone, the mean period of follow-up, at 10.3 ± 9.1 months, was shorter than for those treated with acetylsalicylic acid group, at 17.9 ± 13.7 months, and also shorter than for those receiving no treatment, who had been followed for 28.5 ± 13.5 months (p < 0.05). In 23 of the patients (57.5%), regurgitation disappeared within this period. Apart form one patient, the extent of valvar lesions in the remaining patients did not change throughout the period of follow up. In the outstanding patient, our second during the period of study, the extent of valvar pathology improved but did not disappear. The regurgitation disappeared in 7 (41.2%) of the 17 patients treated with acetylsalicylic acid, in 9 of the 13 (69.2%) treated with prednisone, and in 7 (70%) of the 10 patients who received no anti-inflammatory treatment. The rates of resolution of the valvar lesion among these groups were similar (p > 0.05). The mean duration of the resolution of valvar regurgitation was 12.7 ± 13.2 months in patients treated with acetylsalicylic acid, 4.4 ± 4.8 months in those treated with prednisone, and 21.7 ± 15.5 months in the patients who had no treatment. The patients treated with prednisone were the fastest to recover (p < 0.05) (Table 2).

Table 2. Characteristics of the patients whose valvitis resolved.

Discussion

Carditis is, potentially, the most severe manifestation of acute rheumatic fever, because it is the only one which results in significant sequels and death. Marked clinical evidence of cardiac involvement results in a more characteristic presentation of the disease. On the other hand, the disease may present a more challenging picture when carditis is mild. Since the clinical findings of carditis are mostly represented by the auscultatory findings of valvitis, it is exceedingly difficult to identify minor degrees of abnormalities when using only clinical methods. If there are no clinical findings, therefore, pathologic valvar regurgitation, if detected echocardiographically, should be considered as subclinical carditis.⁹

It is well established that Doppler echocardiography is more sensitive in demonstrating minor degrees of valvar regurgitation than clinical examination. Our present investigations confirmed our previous findings, and those of others, in showing that the addition of Doppler echocardiography gave useful diagnostic information in the setting of silent carditis.^{2, 5, 8, 10} Steinfeld et al., in 1986,¹⁰ were the first to describe silent carditis, showing that of 14 children with acute rheumatic fever, five had only Doppler evidence of mitral regurgitation in the absence of any audible murmur. More recently, in a large cohort of patients with rheumatic fever, it was shown that Doppler echocardiography revealed cardiac involvement in almost nine-tenths of patients during the acute phase, with one sixth of the total having silent carditis.¹¹

In the face of these clinical findings, some now suggest that subclinical valvitis be considered as a minor criterion in diagnosis of acute rheumatic fever, in this way avoiding over-diagnosis,^{2, 7} while others advocate that it should be accepted as a major criterion.^{5, 8, 12} The obvious concern is that physiologic regurgitation may be misinterpreted as abnormal or pathologic, resulting in an over-diagnosis of rheumatic fever.¹³ In this respect, Yoshida et al.¹⁴ have found physiologic mitral regurgitation to be present in almost half the healthy subjects they examined. On the other hand, other studies have reported a very low frequency of physiological mitral regurgitation in children and adolescents, at less than 3%, and found aortic regurgitation to be very rare.^{15, 16} Furthermore, physiologic mitral regurgitation typically produces laminar reversed flow, with the colour jet localized to the vicinity of the mitral valve, and being of shorter duration, occupying less than 60% of systole.⁷ In our study, we have used strict criterions for the echocardiographic diagnosis of pathologic valvitis. In this respect, Folger et al.² detected mitral regurgitation in 10 of 11 cases with acute rheumatic fever, but using similar criterions found no evidence of silent mitral regurgitation in healthy children. Thus, the pathologic and silent mitral regurgitation of rheumatic fever can be distinguished from physiologic mitral regurgitation using the same Doppler criterions, particularly when the jet is directed posteriorly.⁷ In all our patients, the colour jet produced by the mitral regurgitant jet was directed posteriorly.

During the period of follow-up, we found that 23 of our 40 patients with silent carditis became free of valvar findings. Valvar regurgitation persists in 17 patients, and they are still undergoing follow-up. Since the colour Doppler findings have returned to normal, we believe these findings are pathological. Lack of an accurate diagnosis of cardiac lesions can cause under-diagnosis of the disease, and interfere in correct management. Thus, Figueroa et al.¹² suggested that silent carditis is not necessarily a transient entity, because valvar disease persisted after five years of follow-up in at least three-fifths of their cases, despite continuous prophylaxis with penicillin, and no evidence of recurrent disease. We agree, therefore, that subclinical valvitis should now be accepted as evidence of carditis, and thus become a major diagnostic criterion for acute rheumatic fever.

Anti-inflammatory medication is required for carditis, but not for pure chorea. Valvar disease is more frequent with chorea, with Veasy et al.³ demonstrating silent valvar regurgitation in patients with only chorea by colour Doppler echocardiography, and Elevli et al.⁴ finding silent carditis in almost half of their 22 patients with pure chorea. In our study, 3 of 10 patients presenting with chorea still have valvar regurgitation after a mean period of follow up of 18.1 ± 13.9 months. It is important to remember that, when prophylaxis is not given to patients with silent carditis, the valvar lesions may worsen with recurrent attacks of carditis. Because of this, we think that prophylaxis should be continued longer in patients with silent carditis, pure chorea, or isolated arthritis, basing the duration of prophylaxis on echocardiographic examination.

Currently, there is limited data concerning the prognosis of patients with subclinical valvitis. Folger et al.² found silent carditis in 16 of their 21 patients, and followed six of these for up to 36 months. Two became completely normal, but four continued to have incompetence of at least one valve. Figueroa et al.¹² found subclinical valvitis in 10 of their 25 patients, but only three retained evidence of subclinical lesions after five years of follow up. In our previous study, we followed 14 patients for a mean period of 4.52 months, and 4 of them became completely normal.⁵ In this study, where our period of follow-up was 18.1 plus or minus 13.9 months, we found that valvar regurgitation had disappeared in just over half. Obviously, the longer the period of follow-up, the greater are the chances of resolution of the valvar lesions. We found no significant differences in the disappearance of valvar regurgitation according the mode of treatment, failing to show any superiority of prednisone over acetylsalicylic acid. The patients treated with steroids, nonetheless, were treatment with steroids, albeit that those treated with prednisone were significantly the fastest to lose evidence of valvar regurgitation. Recently, we have given prednisone over a shorter time to our patients with silent carditis. For this reason, the patients treated with prednisone have undergone a shorter mean period of follow-up, and the shortness of this follow-up could have affected our results. If we follow these patients treated with prednisone over a longer period, more might prove to be free of valvitis.

Doppler echocardiography, therefore, has become a crucially important tool for the evaluation of cardiac involvement in patients with acute rheumatic fever, both in identifying valvar lesions and assessing the degree of cardiac involvement. We contend that subclinical valvitis demonstrated by echocardiography should now be accepted as adequate evidence for the diagnosis of carditis, and should become a major diagnostic criterion for acute rheumatic fever. The prognosis of silent carditis, and the need for prophylaxis in its presence, is not yet fully understood. These aspects will only be clarified by long-term prospective follow-up.