A 62-year-old man presented with a 10-month history of dyspnea after COVID-19 infection. Dyspnea became worse in the month preceding presentation. Ten months earlier, he had taken a trip to Australia to visit his daughter, where he became infected with COVID-19. He had a fever and a slight cough. He had difficulty in breathing when he took a 30-minute walk. He took over-the-counter medication for a common cold; this did not include paxlovid or cortisone. The symptoms of fever and cough lessened. However, there was no reduction in the dyspnea; he sought no medical care for this symptom. For the one month preceding presentation, he found difficulty in breathing after only a 10-minute walk. There was no fever, cough or sputum. The patient’s medical history included twenty years of hypertension. He took valsartan and hydrochlorothiazide capsules 125 mg per day and amlodipine besylate 5 mg per day. The patient had given up smoking and drinking for one year. He had no pets. Two years ago, he retired, and had had no contact with any toxic environment.

His vital signs were within normal limits. Oxygen saturation was 89% at room air. Auscultation revealed crackles in the lower lung fields. No lymph nodes were palpable. Heart and abdomen were unremarkable. The patient denied intermittent fever, rash, dry eye, dry mouth, Raynaud’s phenomenon, arthritis or symmetrical weakness of the extremities. No skin lesions (heliotrope rash, Gottron’s papules, shawl sign, mechanical hands) were found.

Chest imaging, spirometry and blood examination were ordered to assess parenchymal lung disease. White-cell count was 10 × 103/µL (neutrophils 7.15 × 103/µL, lymphocytes 1.76 × 103/µL, and eosinophils 0.35 × 103/µL). Hemoglobin level was 12.9 g per deciliter; platelet count was 400 × 103 /µL; C-reactive protein level was raised (31.8 mg/L). Liver enzymes and creatinine levels were normal. The patient had a restrictive defect on spirometry. The forced expiratory volume in one second (FEV1) was 61.4% of predicted. The forced vital capacity (FVC) was 51.2% of predicted. The FEV1/FVC ratio was 81.56%, i.e., in the normal range. The total lung capacity (TLC) was 62.3% of predicted. The diffusing capacity of the lung for carbon monoxide (DLCO) was low, 42.1% of predicted.

The chest CT of the patient presented with a complex constellation of features (Fig. 1). The infiltration as seen on CT consisted of two components: multifocal, subpleural, bilateral opacities; and the infiltration around the bronchovascular bundle in the bilateral lower lung field. Some individuals who had been infected with COVID-19 could experience long-term effects from their infection, known as long COVID. Empirically, the subpleural bilateral opacities were interpreted as being due to long COVID [14].

Fig. 1
figure 1

AB, chest CT scan revealing multiple multifocal, subpleural, bilateral opacities (black arrow, A) and infiltration around the bronchovascular bundle in the bilateral lower lung field (white arrow, B)

With regard to the cause of the infiltration around the bronchovascular bundle, the following observations were made. The sample from the patient’s mouth tested negative for COVID-19 using PCR. He tested positive for the antibody, RO-52+, EJ+. Transbronchial cryobiopsy (TBCB) was performed on the right lower lobe posterior segmental bronchus. The biopsy reported chronic inflammation (mainly interstitial pneumonia). T. whipplei was the sole possible pathogen detected in the bronchoalveolar lavage fluid(BALF)(670 reads) by next-generation sequencing technology (NGS). BALF contained 800 cells/µL (64% lymphocytes, 22% neutrophils, 10% macrophages, 4% eosinophils).

The concentrations of skeletal muscle enzymes were normal. He was assessed by a rheumatologist and a presumptive diagnosis of connective tissue disease-interstitial lung disease (CTD-ILD) was made.

The significance of T. whipplei detection in BALF still needed to be considered, in terms of whether this involved a carrier or a pathogen. There was no fever, cough or sputum. The pulmonologists initially concluded that T. whipplei was colonized in BALF. Prednisone and cyclophosphamide were given.

At follow-up one month later, the patient complained that the dyspnea had only improved a little. He had a low fever 3 days earlier. Oxygen saturation was still low, at 89%. The C-reactive protein level was 28.0 mg/L. A CT scan of the chest showed partial absorption of the opacities. Most of the subpleural opacities due to long COVID resolved after prednisone treatment. New diffuse ground-glass infiltration along bronchovascular bundles in bilateral lung field occurred (Fig. 2). The recent fever and the new diffuse infiltration on CT suggested an infection.

Fig. 2
figure 2

Chest CT at initial diagnosis is shown in A, C and E. Chest CT scan after prednisone treatment for one month (B, D, F) revealed partial absorption of the opacities. New diffuse ground-glass infiltration along bronchovascular bundles (white arrow)in bilateral lung field

The clinical suspicion for the earlier diagnosis of CTD-ILD therefore had to be re-considered. The pathogen in the BALF, T. whipplei, was then considered as a potential contributing factor. On pathological re-evaluation of the TBCB specimen, foamy macrophages in the alveolar air space were found, and periodic acid-Schiff (PAS) staining revealed numerous intracytoplasmic organisms, with the morphologic features consistent with T.whipplei (Fig. 3). Lung T. whipplei infection was therefore considered responsible for the symptoms in this case. Cyclophosphamide was discontinued and the dosage of prednisone rapidly reduced. The patient received intravenous ceftriaxone for two weeks. Oral trimethoprim–sulfamethoxazole at a dose of 160 mg of trimethoprim and 800 mg of sulfamethoxazole twice daily was prescribed.

Fig. 3
figure 3

Hematoxylin and eosin staining (A: low-power photomicrograph; B: original magnification×100) showing the interstitial pneumonia with lymphocyte and plasmacyte infiltration of the alveolar septum and focal proliferation of fibrous tissue. Numerous foamy macrophages are seen in the alveolar air space. The foamy macrophages stained positive for CD68 by immunohistochemistry (C). Periodic acid-Schiff staining showing numerous intracytoplasmic organisms (black arrows), with morphologic features consistent with T. whipplei (original magnification× 200 (D)

One month later, with a gradually reducing dose of prednisone to 5 mg per day, the patient returned to the pulmonology clinic and reported an obvious resolution of dyspnea. Oxygen saturation was raised to 93%, and the CT scan showed only a few fibrous cords and reticular opacities (Fig. 4). C-reactive protein level was normal, at 4.5 mg/L, and the erythrocyte sedimentation rate was lower than before (24 mm/hr). The improvement of the restrictive lung function defect was demonstrated. The FEV1 was 67.7% of predicted. The FVC was 66.5% of predicted. The TLC was 66.5% of predicted. The DLCO was 50.7% of predicted. The trimethoprim-sulfamethoxazole treatment was continued; the prednisone discontinued.

Fig. 4
figure 4

The CT scan after antibiotic administration (B, D) shows the diffuse infiltration was partially absorbed. The lung field was clearer than before. Chest CT scan before the antibiotics is showed in A and C

Two months later, in the pulmonology clinic, the patient reported complete relief of his earlier breathing difficulties; he could routinely take a walk of several kilometers per day.

Source link