Cytokine levels in sputum, not serum, may be more helpful for indicating the damage in the lung and the prognosis of severe COVID-19 – A case series

Abstract

Purpose

To describe the relationship between the severity of lung damage and cytokine levels in sputum, bronchoalveolar lavage fluid (BALF), serum.

Method

Eight severe patients infected with coronavirus disease 2019 (COVID-19) were admitted and their cytokines and chest computed tomography (CT) were analyzed.

Results

Compared with in serum, IL-6 and TNF-α in sputum and in BALF show more directly reflect the severity of COVID-19 critical patients. The gradient ratio of IL-6 levels may predict the prognosis of severe patients.

Conclusion

Cytokine levels in the sputum may be more helpful for indicating lung damage. Local intervention through the respiratory tract is expected to benefit patients with severe COVID-19.

Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection can cause cytokine storm and acute respiratory distress syndrome (ARDS) or even death. IL-6 and TNF-α levels in serum are independent and important predictors of coronavirus disease 2019 (COVID-19) severity and death¹. Our research finds that IL-6 and IL-10 are involved in the deterioration of COVID-19. IL-6 levels in bronchoalveolar lavage fluid (BALF) and in the pleural effusion of COVID-19 patients are significantly higher than that in peripheral blood². We aim to explore the relationship between the severity of lung damage and cytokine levels in sputum, BALF and serum. Here we analyze eight severe patients with SARS-CoV-2.

Methods

From January 2021 to March 2021, eight severe COVID-19 patients were admitted and evaluated at COVID-19 Critical Care Center in Heilongjiang Province, China. According to the treatment guidelines of China's national COVID-19 intensive care expert group, all patients received oxygen therapy, expectoration, daily prone ventilation for 12–20 h, anti-infection, anticoagulant, and symptomatic treatment. Serum samples were collected, and bronchoscopy was performed within 24 h after mechanical ventilation. During the process of bronchoscopy, sputum samples from both lungs were taken, and bronchoscopy alveolar lavage was performed. Cytokine levels were detected by ELISA. Chest computed tomography (CT) examination was performed within 24 h before or after mechanical ventilation.

Results

The median age of the eight patients is 72 years (range 54–79 years) and the median intensive care unit (ICU) hospitalization days is 40.5 days (range 29–66 days). Seven of them have chronic diseases. The most common symptoms before admission are fever (N = 7), fatigue (N = 5), cough (N = 5), sputum (N = 3), and pharyngalgia (N = 3); medium common symptoms are dyspnea (N = 2), dizziness (N = 2). The patients usually take one or two weeks to become severe after the symptom's onset. Three of them had thrombocytopenia (#1, #2, #5). All of the cases require supplemental oxygen and underwent high-flow oxygen inhalation or non-invasive mechanical ventilation, and finally mechanical ventilation. Except for #4, all patients had a tracheotomy.

Some patients received antibody plasma (#8), glucocorticoid therapy (#1), continuous renal replacement therapy (CRRT) (#7, #8), and extracorporeal membrane oxygenation (ECMO) (#1, #4, #7, #8). So far, four patients have finally recovered, and four patients died (#4, #5, #7, #8), but #8 was died of tumor-related complications after treating COVID-19 successfully and removing the ventilator (Table 1 ).

Table 1.

Patient characteristics and treatment.

	Patient
Characteristics	#1	#2	#3	#4	#5	#6	#7	#8
Sex	F	M	F	M	M	M	M	M
Age, y	72	76	66	54	77	72	79	74
Comorbidity	HT, DM, AF, CI, CH	AF, HT, CI	HT	None	HT, COPD	CI, Lung carcinoma, Chronic bronchitis	HT, DM, CI	HT, DM, Renal carcinoma, Lung metastases, Emphysema
Symptom	Fatigue, Dyspnea	Fever, Fatigue	Fever, Cough, Fatigue, Pharyngalgia, Dizziness	Fever	Fever, Cough, Sputum, Pharyngalgia, Dizziness	Fever, Cough, Sputum	Fever, Cough, Fatigue	Fever, Cough, Sputum, Fatigue, Pharyngalgia
Severity on diagnosis of pneumonia	Severe	Severe	Moderate	Moderate	Moderate	Moderate	Moderate	Severe
Disease presentation and course
Interval between symptom onset and ICU admission	5	6	10	7	18	11	15	10
ICU hospitalization days	48	29	33	42	62	39	37	66
Treatment
Steroids	Methylprednisolone	None	None	None	None	None	None	None
Convalescent Plasma	None	None	None	None	None	None	None	400ml
Etesevimab (LY-CoV016, 600 mg, 6 ml)	30ml	30ml	29ml	None	34ml	40ml	30ml	42ml
Other	ECMO	None	None	ECMO	None	None	ECMO&CRRT	ECMO&CRRT
Outcome	Recovered	Recovered	Recovered	Death	Death	Recovered	Death	Death

F, female; M, male; HT = hypertension; DM = diabetes mellitus; AF = atrial fibrillation; CI = cerebral infarction; CH = cerebral hemorrhage; ICU = intensive care unit; CRRT = continuous renal replacement therapy; ECMO = extracorporeal membrane oxygenation.

The levels of IL-6, IL-10, and TNF-α in sputum and in BALF from both lungs were detected. The sputum sample of patient #3 and the sputum sample from the left bronchus of patient #8 was not obtained during bronchoscopy. Except for #1, IL-6 and TNF-α levels in sputum are slightly higher than that in BALF with small differences but significantly higher than that in serum. The IL-6 levels in BALF and in sputum of three patients who died of COVID-19 are 4.63 to 27.16 times and 13.06 to 37.82 times higher than that in serum, respectively (#4, 4.63 & 13.06; #5, 13.42 & 25.05; #7, 27.16 & 37.82). Levels of IL-6 in BALF and in sputum of COVID-19 recovered patients are 0.57 to 8.84 times and 0.32 to 4.72 times higher than that in serum (#1, 0.57 & 0.32; #2, 1.76 & 2.88; #3, 8.84 & NA; #6, 3.28 & 4.72; #8, 1.33 & 2.43) (Table 2 ). In addition, levels of IL-6 and TNF-α in sputum and in BALF are related to the severity of lung injury. Except for #7 (CT shows both lungs are extensively affected, but cytokine levels of the left lung are higher than that of the right lung), the cytokine levels in sputum from infected lung (left/right) are consistent with the severity of the corresponding CT imaging. Compared with in sputum, IL-6 and TNF-α levels in BALF from #4 and #5 show some deviation. (Table 3 , Fig. 1 )

Table 2.

Clinical characteristics and Laboratory findings.

	Patient
Variables	#1	#2	#3	#4	#5	#6	#7	#8
Clinical characteristics
Body temperature, °C	37.2	37.3	37.3	37.6	38.1	37.8	36.9	37.4
PaO2/FiO2	100	167.2	168	117.2	216.3	152	123.6	82.7
Mechanical ventilation, d	26	16	15	36	59	27	33	63
ECMO, d	16	NA	NA	24	NA	NA	20	21
Laboratory findings
Serum
IL-6 (1.18–5.3pg/ml)	92.65	2512.21	85.06	311.76	300.11	323.98	197.09	765.86
IL-10 (0.19–4.91pg/ml)	3.78	27.19	6.63	16.34	9.38	7.35	3.23	9.63
TNF-α (0.1–2.31pg/ml)	1.43	1.04	1.5	1.18	1.95	1.37	0.84	0.8
C-reactive protein  (0–10 mg/L)	96.85	234.64	52.85	242.68	227.04	228.96	122.38	180.73
Sputum
IL-6 (1.18–5.3pg/ml)
Left	5.69	7227.85	NA	889.08	7517.66	777.46	7454.47	NA
Right	29.90	1907.96	NA	4072.92	3113.14	1529.61	5111.11	1859.87
TNF-α (0.1–2.31pg/ml)
Left	12.63	244.02	NA	73.78	21.71	6.58	712.30	NA
Right	213.55	40.94	NA	286.76	17.88	37.7	421.76	127.26
BALF
IL-6 (1.18–5.3pg/ml)
Left	44.13	4434.00	373.30	1444.92	1473.43	367.08	5352.55	363.26
Right	52.76	2518.69	751.52	411.39	4026.66	1062.52	4750.29	1018.25
TNF-α (0.1–2.31pg/ml)
Left	17.10	123.05	55.39	87.38	10.22	16.55	92.55	4.19
Right	17.96	66.56	136.86	26.93	15.82	28.38	29.38	33.53

BALF = bronchoalveolar lavage fluid; ECMO = extracorporeal membrane oxygenation.

Table 3.

The cytokines level in Sputum and BALF associated with CT images.

	Patient
Variables	#1	#2	#3	#4	#5	#6	#7	#8
IL-6 and TNF-α in Sputum
Left	L	H	NA	L	H	L	H	NA
Right	H	L	NA	H	L	H	L	H
IL-6 and TNF-α in BALF
Left	L	H	L	H	L	L	H	L
Right	H	L	H	L	H	H	L	H
CT
Left/Right serious	Right	Left	Right	Right	Left	Right	Both	Right

BALF = bronchoalveolar lavage fluid; CT = computed tomography; L= lower, H = higher, compare with the contralateral lung.

Fig. 1.

CT imaging of eight COVID-19 patients.

Discussion

The systemic levels of cytokines caused by COVID-19 may be lower than sepsis, but the local response is more intense³. Cytokine levels in circulation may not accurately reflect that in local tissue. The samples from BALF have the highest positive rate of detecting SARS-CoV-2, followed by sputum samples from the lower respiratory tract⁴. Most COVID-19 patients are accompanied by elevated serum IL-6 before their lung lesions becoming worse, and the consistently high IL-6 level may be a risk factor for persistent lung injury⁵. The lung is the core site of SARS-CoV-2 infection that may cause a different mechanism from the other cytokine storm-related diseases. Recently, the intervention methods through the respiratory tract attract more attention. In one animal experiment, intranasal administration of specific dimer lipopeptides can effectively block the host SARS-CoV-2 virus replication⁶. And to inhale CD24 protein by exosomes into the lungs may inhibit the cytokine storm⁷.

Conclusion

The inflammatory response plays a key role in COVID-19 and the cytokine storm aggravates the severity of patients. Compared with in serum, IL-6 and TNF-α in sputum and in BALF may be more directly reflect the severity of COVID-19 critical patients. Cytokine levels in the sputum may be helpful for indicating the extent of damage in the lung and IL-6 may predict the prognosis of patients with severe COVID-19. Local intervention through the respiratory tract may be a reasonable treatment for high-risk patients with severe COVID-19.

Funding

This study supported by Dr. Wang's grant: Provincial Natural Science Foundation Outstanding Youth Project (NO. JQ2021H003).

Availability of data and materials

The datasets used and/or analyzed during this study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

Human research ethics committee approval for the study was provided by local institution. Patients informed consent.

Not applicable.

Consent for publication

Not applicable.

CRediT authorship contribution statement

Changsong Wang: Methodology. Kai Kang: Methodology. Xiuwen Lan: Writing – original draft. Dongsheng Fei: Writing – original draft. Qian Wang: Writing – original draft. Xianyong Li: Writing – review & editing. Yang Chong: Writing – review & editing. Yan Gao: Writing – review & editing. Huaiquan Wang: Writing – review & editing. Xueting Li: Writing – review & editing. Mingyan Zhao: Writing – review & editing. Kaijiang Yu: Writing – review & editing.

Declaration of Competing Interest

None of the authors reports any competing interests.