Organizing Pneumonia in a Case of Cold Agglutinin Disease with Pulmonary Thrombosis

Abstract

Cold agglutinin disease is a subtype of autoimmune hemolytic anemia that occurs via the activation of specific anti-red blood cell antibodies (agglutinins) at low temperatures. Autoimmune hemolytic anemia has been reported to cause interstitial pneumonia; however, the underlying mechanism remains unclear. We herein report a 46-year-old man diagnosed with cold agglutinin disease complicated by pulmonary thrombosis and organizing pneumonia. Treatment with prednisolone improved the course of cold agglutinin disease and organizing pneumonia in a similar manner. To our knowledge, this is the first report of cold agglutinin associated with organizing pneumonia, suggesting a potential link between the two.

Introduction

Autoimmune hemolytic anemia (AIHA) is an autoimmune disease characterized by the production of anti-red blood cell (RBC) antibodies. Depending on the presence or absence of complement activation, AIHA can cause hemolysis or anemia, respectively (1,2). Anti-RBC autoantibodies generated by self-reactive B lymphocytes attack erythrocytes, a process known as “opsonization,” which accelerates the phagocytosis of macrophages (2). Furthermore, complement activation not only induces the membrane attack complex (MAC) and antibody-dependent cell-mediated cytotoxicity (ADCC) but also increases phagocytosis, resulting in the exacerbation of hemolysis (2).

AIHA is an uncommon disease with a morbidity rate of 0.8-3% in adults (3). Although half of AIHA cases are regarded as idiopathic, the remaining half are secondary to lymphoproliferative disease, chronic infection, autoimmune disease, or tumor (1,3). AIHA can be classified into subtypes according to autoantibody immunoglobulin (Ig) classes and the optimal activation temperature, namely warm, cold, and mixed types. Cold agglutinin disease (CAD), known as the cold type, accounts for 15% of AIHA cases and is mediated by specific autoantibodies [cold agglutinins (Cas)], mainly IgM, IgG, and IgA (1,4). At an optimal temperature of 3-4°C, Cas are activated and agglutinate RBCs (4,5). Although CAD chiefly causes anemia and circulatory symptoms, such as acrocyanosis and Raynaud phenomena, it was recently reported that CAD could be a risk factor for thrombosis (5,6).

Organizing pneumonia (OP) is defined as interstitial pneumonia (IP) caused by the inflammatory process of repairing injured pulmonary tissue (7). During this process, fibrin exudates and loose collagen-containing fibroblasts accumulate in the distal air space and sometimes in the bronchial lumen, thereby developing buds of granulation tissue (8). OP can be cryptogenic in origin or secondary to a tumor, infection, or systemic autoimmune disorder, such as collagen disease (7,8).

We herein report a case of CAD associated with pulmonary thrombosis (PT) and OP. To our knowledge, there are no other reports of CAD accompanied by PT and OP.

Case Report

The patient was a 46-year-old man with no significant medical history or transfusion records. In July 2021, he suddenly experienced chest pain; however, he did not visit the hospital and the pain disappeared within a day. Two months later, during recurrence of chest pain, he received electrocardiogram and echocardiogram examinations at a local hospital, but no abnormal findings were detected. Similar to the previous time, the pain improved immediately. After a month, he developed dyspnea with chest pain; therefore, he visited our hospital and underwent examinations.

On presentation, jaundice and cicatricial purpura in both forearms were confirmed. An arterial blood gas examination showed the following: pH 7.497, PaCO₂ 33.2 mmHg, PaO₂ 73.4 mmHg, and HCO₃^- 25.1 mEq/L. A blood examination revealed some notable abnormal findings (Table), including a low RBC count of 2.20×10⁶/μL, hemoglobin (Hb) 7.6 g/dL, nearly normal mean corpuscular volume 100.5 fL, and mean corpuscular hemoglobin concentration (MCHC) 34.4 g/dL indicated normocytic normochromic anemia. In addition, he showed a high reticulocyte count of 75.0 ‰, total bilirubin 3.9 mg/dL, indirect bilirubin (I-Bil) 3.6 mg/dL, and low haptoglobin 4.0 mg/dL implied hemolytic anemia. He also showed high IgM 315 mg/dL, low hemolytic complement activity (CH50) 7.0 U/mL, complement C4 4 mg/dL, a positive direct Coombs test, and cold agglutinin reaction titers of 1:1,280 were presented, which suggested CAD. A high white blood cell (WBC) count of 10,300 /μL, dominated by neutrophils 85.9%, C-reactive protein (CRP) 0.58 mg/dL, Krebs von den Lungen-6 (KL-6) level of 509 U/mL, and D-dimer 14.9 μg/mL were also noted. Tumor markers, collagen disease markers, and infectious markers were not detected. Sputum culture and severe acute respiratory syndrome corona virus 2 polymerase chain reaction tests showed no abnormal findings.

Table.

Laboratory Data on Admission.

	Complete blood cell count		Normal range
WBC	10,900	/µL	3,500-9,000
Neut	75.6	%	44-69
Eos	7.3	%	0-5
Lym	13	%	24-44
Baso	0.8	%	0-2
RBC	2.2	×106/µL	4.2-5.6
Hgb	7.6	g/dL	13.5-18.0
MCV	100.5	fL	83.0-101.0
MCH	34.5	pg	28.0-35.0
MCHC	34.4	g/dL	30.5-36.0
Plt	438	×103/µL	125-370
Ret	75	‰	8-22
	Coagulation fibrinolysis examination
PT	12.2	s	9.2-12.8
PT%	91	%	80-
PT-INR	1.05
aPTT	25.6	s	24.5-36.5
Fib	584	mg/dL	168-403
D-dimer	14.9	µg/mL	0.0-1.0
	Electrolyte and blood chemistry
TP	6.7	g/dL	6.5-8.2
Alb	3.9	g/dL	3.4-4.7
T-Bil	3.9	mg/dL	0.2-1.0
D-Bil	0.3	mg/dL	0.1-0.3
I-Bil	3.6	mg/dL	0.1-0.7
UA	7.1	mg/dL	3.4-7.0
BUN	11.7	mg/dL	6.0-20.0
Cre	0.7	mg/dL	0.6-1.2
eGFR	95.5	mL/min/1.73 m2	60-100
AST	24	U/L	7-38
ALT	9	U/L	8-43
LD	389	U/L	124-222
γ-GTP	223	U/L	0-50
ALP	222	U/L	38-113
CRP	0.58	mg/dL	0.00-0.30
Sodium	140	mmol/L	134-147
Potassium	4.4	mmol/L	3.7-5.0
Chloride	104	mmol/L	96-110
Serum iron	120	µg/dL	80-150
TIBC	131	µg/dL	290-400
Ferritin	369	ng/mL	25.0-232.0
IgG	1,039	mg/dL	917-2,011
IgA	168	mg/dL	106-442
IgM	311	mg/dL	44-287
C3	72	mg/dL	50-98
C4	2	mg/dL	18-49
CH50	7	CH50U/mL	23.0-46.0
Cold agglutination reaction	1,280	titer
Direct Coombs test	Positive
RF	5	IU/mL	0-15
ANA	40>	titer	0-40
Anti SS-A Ab	1>	U/mL	0.0-9.9
Anti SS-B Ab	1.3	U/mL	0.0-9.9
C-ANCA	1>	U/mL	0.0-3.4
P-ANCA	1>	U/mL	0.0-3.4
sIL-2R	670	U/mL	122-496
Haptoglobin	4	mg/dL	41.0-273.0
KL-6	509	U/mL	0-500
HTLV-1 Ab	0.1	C.O.I	0.00-0.99
HIV Ag/Ab	0.3	C.O.I	0.0-0.9

ALT: alanine aminotransferase, Alb: albumin, ALP: alkaline phosphatase, ANA: anti-nuclear antibody, Anti-SS-A Ab: anti-SS-A/Ro antibody, Anti-SS-B Ab: Anti-SS-B/La antibody, aPTT: activated partial thromboplastin time, AST: aspartate aminotransferase, Baso: basophil, BUN: blood urea nitrogen, C-ANCA: cytoplasmic antineutrophil cytoplasmic antibody, CH50: 50% hemolytic complement, C.O.I: cut off index, Cre: creatinine, CRP: C-reactive protein, C3: complement 3, C4: complement 4, D-Bil: direct bilirubin, Eos: eosinophil, eGFR: estimated glomerular filtration rate, Fib: fibrinogen quantity, γ-GTP: gamma-glutamyl transpeptidase, Hgb: hemoglobin, HIV Ag/Ab: human immunodeficiency virus antigen/antibody, HTLV-1 Ab: human T-lymphotropic virus type I antibody, I-Bil: indirect bilirubin, IgA: immunoglobulin A, IgG: immunoglobulin G, IgM: immunoglobulin M, INR: international normalized ratio, KL-6: Krebs von den Lungen-6, LD: lactate dehydrogenase, Lym: lymphocyte, MCH: mean corpuscular hemoglobin, MCHC: mean corpuscular hemoglobin concentration, MCV: mean corpuscular volume, Neut: neutrophil, P-ANCA: perinuclear antineutrophil cytoplasmic antibody, Plt: platelet, PT: prothrombin time, RBC: red blood cell, Ret: reticulocytes, RF: rheumatoid factor, sIL-2R: soluble interleukin-2 receptor, T-Bil: total bilirubin, TIBC: total iron binding capacity, TP: total protein, UA: uric acid, WBC: white blood cell

Further examinations were performed to enable a precise diagnosis. On a bone marrow examination, only erythroid hyperplasia, no abnormal cells or morphologies, was observed. Chromosome testing revealed no abnormalities. Flow cytometry of the bone marrow was performed to rule out malignant lymphoma and chronic lymphocytic leukemia. Lymphocyte gating in the bone marrow showed a slight increase in B lymphocytes and plasma cells. Light-chain gating showed a slightly increased percentage of Kappa and Gamma at 3.5. These findings were consistent with nonspecific inflammatory changes and were not suggestive of malignant lymphoma or chronic lymphocytic leukemia.

Chest radiography (CXR) revealed abnormal shadows in the bilateral upper lung fields (Fig. 1). Computed tomography (CT) and angiography of the chest region showed airspace consolidations predominantly in the bilateral upper lobes and filling defects, indicating pulmonary thrombosis (PT) in the pulmonary artery of the right lung, while splenomegaly was detected in the abdominal region (Fig. 2). Pulmonary perfusion scintigraphy (PPS) demonstrated scattered bilateral defects in lung perfusion (Fig. 3).

Figure 1.

Chest radiography showing abnormal shadows in the bilateral upper lung fields. The shadow occupies the entire area of the right upper lung (red arrows) and the mediastinal side of the left upper lung (blue arrows).

Figure 2.

Chest computed tomography (CT) revealing abnormal consolidations with air bronchogram in the bilateral upper lobes predominantly, (red arrows) pulmonary window setting in the left two images, and filling defects in the pulmonary artery by angiography in the right upper lobe (blue arrow) in the right upper image. On abdominal CT, splenomegaly is highlighted by the green arrows in the right lower image.

Figure 3.

Pulmonary perfusion scintigraphy showing defects of perfusion scattered in the bilateral lung (red arrows).

Pulmonary thromboembolism was diagnosed based on high D-dimer levels, a defect in the right inferior pulmonary artery on CT angiography, and a blood flow defect on PPS in an area consistent with the CT angiographic findings.

To investigate air space consolidation, a transbronchial lung biopsy (TBLB) and bronchoalveolar lavage (BAL) by bronchoscopy were conducted from the right upper lobe (B²). BAL showed high WBC counts of 700 /μL, predominantly lymphocytes, 56%; eosinophils, 5.3%; neutrophils, 5.6%; macrophages, 33%; monocytes, 0%; and CD4/8, 0.6. Pathologically, the TBLB showed lymphocyte infiltration into the alveolar interstitium, accumulation of collagen and fibroblasts, and a developing polyp in the alveolar air space (Fig. 4). No notable bacteria were found in the BAL cultures. In addition to the CT findings, BAL and pathological findings of biopsied specimens implied OP. A surgical lung biopsy was not recommended because of the serious anemia and PT.

Figure 4.

Pathological findings of the specimen obtained by a transbronchial lung biopsy. Left: Hematoxylin and Eosin staining shows lymphocyte infiltration and the accumulation of collagen and fibroblasts in the alveolar interstitium and airspace, forming a polyp, indicated by red arrows (×100 magnification). Right upper: Elastica van Gieson staining shows elastic fibers colored blackish brown (×200 magnification). Right lower: Masson Trichrome staining shows collagen colored blue (×200 magnification).

Based on the above results, the patient was diagnosed with CAD complicated by PT and OP. As treatment, 1 mg/kg/day of prednisolone (PS) and 12,000 units (U)/day of heparin were administered. With monitoring of activated partial thromboplastin time, the heparin dose was increased to 28,000 U/day. Gradually, the Hb and D-dimer levels improved, and simultaneously, the volume of the abnormal shadow in the CXR was reduced. Based on his condition, PS and heparin were decreased, and heparin was switched to apixaban.

Two months later, the near disappearance of the PT and OP was confirmed by chest CT (Fig. 5). Control of PS and other medications was continued, and sutimlimab was started during follow-up visits. The patient's condition did not worsen within two years of the start of treatment.

Figure 5.

CT findings before and after treatment with prednisolone and anticoagulant. The bilateral consolidations and filling defects in the pulmonary artery (blue arrows) disappeared after treatment.

Discussion

We herein report a case of CAD complicated by PT and OP. Seventeen cases of AIHA with IP have been reported (9-20). While no association between AIHA and IP has yet been confirmed, some authors have suggested that AIHA could be a cause of secondary IP as a systematic autoimmune disease (10,11). Scadding hypothesized that erythrocytes with an immune complex consisting of abnormal autoantibody-stimulated phagocytosis of circulating macrophages, and macrophages in the alveolar walls results in hemolysis (10). During this process, the inflammatory reaction is enhanced in the alveolar walls, which may eventually induce IP (10). Accordingly, in our case, Cas may also play a role in the mechanism by which CAD induces OP by activating inflammation of alveolar walls.

Of the 17 cases, almost all cases of IP were regarded as usual interstitial pneumonia, nonspecific interstitial pneumonia, or unclassified, while OP was suspected of being bronchiolitis obliterans organizing pneumonia in one report (16). That case was similar to ours in terms of the concurrent onset between AIHA and OP, good steroid response to both lymphocyte predominance and low CD4/8 in BAL, and infiltration lymphocytes in alveolar walls evident in biopsied specimens (16). Whether or not IP can be induced by CAD is unclear; therefore, additional cases must be identified.

Two studies found increased PT in patients with CAD compared to non-CAD patients (6,21). In a prospective cohort study conducted in the United States, 4.9% of CAD patients had PT compared with 1.5% of non-CAD patients (6). In a Japanese retrospective study, 3.5% of CAD patients had PT compared with 0.5% of non-CAD patients (21). Although the pathophysiology of thrombosis in CAD remains unclear, it has been suggested that heme released during hemolysis and coagulation produces nitric oxide (NO) scavengers that decrease NO (6,21-23). Furthermore, free heme produces reactive oxygen species, which are released during inflammatory reactions via leukocyte mobilization and cytokine upregulation in monocytes and macrophages (6,23). In contrast, crosstalk between complement activation and coagulation pathways has been suggested to induce thrombosis, independent of hemolysis (21,24). These reactions may cause vasoconstriction, platelet aggregation, and increased expression of endothelial adhesion molecules, resulting in thrombosis (6,21-24).

Several limitations associated with the present study warrant mention. First, there may be an argument that OP coincidentally occurred with CAD; however, given the similar onset, clinical course, and response to PS, it was unlikely that both merely occurred by coincidence. Second, OP might have been caused not by CAD but by PT. Indeed, PT can induce inflammation, so this possibility cannot be excluded. However, we could not find any reports of PT directly causing OP. Third, it is possible that pathogenetic factors common to AIHA and OP, such as autoimmune abnormalities, infections, malignancies, and medications, may have been present in this case (1,5,8). To this end, we investigated such conditions as much as possible, but no other suspicious diseases were identified, even after two years had passed. We therefore believe that there was a direct pathophysiological link between OP and CAD in our case.

Conclusion

We encountered a case of CAD complicated by OP and PT. Our findings suggest that CAD may cause not only PT but also OP; therefore, pulmonary lesions should receive particular attention if CAD is encountered.

The authors state that they have no Conflict of Interest (COI).