Clinical Course of SARS-CoV-2 Infection and Recovery in Lung Transplant Recipients

Abstract

Background:

Reports on outcomes following SARS-CoV-2 infection in lung transplant recipients remain limited.

Methods:

We performed a single-center, observational study of outcomes in lung transplant recipients diagnosed with SARS-CoV-2 between 5/1/2020 – 3/15/2022 that were followed for a median of 123 days. We analyzed changes in spirometry, ALAD incidence, hospitalization, mechanical ventilation needs, secondary infection, and survival.

Results:

In our cohort of 336 patients, 103 developed COVID (27 pre-Delta, 20 Delta and 56 Omicron-era). Twenty-five patients (24%) required hospitalization and 10 patients ultimately died (10%). Among 85 survivors who completed ambulatory spirometry, COVID-19 did not alter change in FEV₁ or FVC over time compared to the preceding 6 months. The pre-COVID FEV₁ change was −0.05 mL/day (IQR −0.50 – 0.60) compared to −0.20 mL/day (IQR −1.40 – 0.70) post-COVID (p= 0.16). The pre-COVID change in FVC was 0.20 mL/day (IQR −0.60 – 0.70) compared to 0.05 mL/day (IQR −1.00 – 1.10) post-COVID (p= 0.76). While the cohort overall had stable lung function, 33 patients (39%) developed ALAD or accelerated CLAD (FEV₁ decline >10% from pre-COVID baseline). Nine patients (35%) with ALAD recovered lung function. Within 3-months of acute COVID infection, 18 patients (17%) developed secondary infections, the majority being bacterial pneumonia. Finally, vaccination with at least 2 doses of mRNA vaccine was not associated with improved outcomes.

Conclusions:

This study describes the natural history of SARS-CoV-2 infection in a large cohort of lung transplant recipients. Although one-third of patients develop ALAD requiring augmented immunosuppression, infection with SARS-CoV-2 is not associated with worsening lung function.

Graphical Abstract

INTRODUCTION:

The novel coronavirus, SARS-CoV-2, has emerged as a leading cause of morbidity and mortality throughout the general population worldwide.¹ Hospitalization rates of those infected range from 0.5% to 9% (depending on age) and the case fatality rate is 1.2%.² Solid-organ transplant recipients are at increased risk of severe COVID-19 illness, with mortality as high as 24%.³ In lung transplant recipients specifically, high hospitalization rates with mortality of 10–39% have been reported.^4–7,31,32 In addition, co-infections and secondary infections are common in the immunosuppressed population.^33,34 There are limited data on risk factors for severe disease in lung transplant patients with SARS-CoV-2 infection. One study showed that patients with pre-existing chronic lung allograft dysfunction (CLAD) had more rapid decline in lung function and higher mortality after COVID-19.⁸ Vaccination against SARS-CoV-2 has been shown to be protective for solid-organ transplant recipients; despite having a limited ability to generate an immune response,^9–12 transplant recipients who received two-doses of mRNA vaccine had reduced mortality from COVID-19.¹³

The long-term effects of COVID-19 on lung allograft function and patient survival are not well understood. Mahan and colleagues reported frequent persistent lung parenchymal abnormalities on imaging (65%), mild to moderately impaired performance status, and a persistent decline in FEV₁ or FVC >10% of baseline in a single-center analysis of 44 patients with median follow-up of 4.5 months post-COVID.¹⁴ In 24 lung transplant recipients with COVID-19, Permpalung et al. showed a high rate of hospital readmissions and secondary bacterial and fungal infections, though did not detect an increased incidence of acute cellular rejection (ACR), antibody-mediated rejection (AMR), de novo donor-specific antibodies (dnDSA), or lung function decline within 60–90 days of infection compared to matched controls.¹⁵

We report clinical outcomes following SARS-CoV-2 infection in an observational single-center study of 103 lung transplant patients across multiple SARS-CoV-2 variants. We focus specifically on the impact of COVID-19 on lung function, risk of acute lung allograft dysfunction (ALAD), and whether decline in lung function during COVID-19 illness is reversible. We hypothesize that acute SARS-CoV-2 infection will lead to an increased rate of lung function decline in lung transplant recipients.

METHODS:

Study population

We performed a single-center, observational study [VUMC Institutional Review Board #210566]. Adult (>18 years) lung transplant recipients who were at least 3 months post-transplant were included. A diagnosis of COVID-19 was based on positive quantitative polymerase chain reaction assays of nasopharyngeal swabs or a positive home test between 3/1/2020 and 3/15/2022. Patients were classified as “pre-Delta variant era” if infection occurred before 5/31/2021, “Delta variant era” if diagnosed between 6/1/2021 and 11/30/2021, and “Omicron variant era” if diagnosed after 12/1/2021, based on epidemiological data of variant prevalence in our geographic region; SARS-CoV-2 strains were not formally genotyped. Patients were considered vaccinated 2 weeks after the second dose of BNT162b2 (Pfizer) or mRNA-1273 (Moderna) vaccines or after 1 dose of JNJ-78436735 (Johnson & Johnson). Monoclonal antibody therapy was bamlanivimab during the pre-Delta era, casirivimab-imdevimab during the Delta era, and sotrovimab during the Omicron era, based on our hospital and regional public health guidelines. Patients were classified as diabetic if hemoglobin A1c level was >6.5% within 3 months of SARS-CoV-2 diagnosis or if they were prescribed glucose-lowering medications. ALAD was defined as a reduction in FEV₁ >10% of pre-COVID-19 values. Secondary infections were diagnosed based on positive culture or PCR data within 90 days of SARS-CoV-2 diagnosis.

Immunosuppressive and antimicrobial protocols

At our center, standard maintenance immunosuppression included a calcineurin inhibitor (tacrolimus preferred, goal trough 10–14 ng/mL within the first-year post-transplant, followed by goal 8–12 ng/mL thereafter), an antiproliferative agent (mycophenolate mofetil preferred, 1000mg twice daily), and prednisone (tapered from 20mg daily to 5mg daily over the first 3 months post-transplant).¹⁶ The anti-proliferative was held if a patient had persistent or recurrent infection, malignancy, or cytopenias. Standard infection prophylaxis included trimethoprim-sulfamethoxazole, valganciclovir (for 6–12 months post-transplant depending on donor/recipient CMV serostatus), and an antifungal (posaconazole preferred through 2 months post-transplant). SARS-CoV-2 vaccination was recommended beginning in January 2021 with mRNA vaccination preferred. Monoclonal antibody therapy was readily available beginning in November 2020.

Management of SARS-CoV-2 infection and recovery

Patients diagnosed with SARS-CoV-2 were managed according to our center’s protocol. Patients were contacted daily for assessment of vital signs and symptoms. Anti-proliferative immunosuppression was held during acute infection. Patients were encouraged to receive monoclonal antibody infusion as soon as possible.^17,18 Patients admitted to the hospital were treated with remdesevir, corticosteroids, and/or baricitinib per hospital protocols at the time of diagnosis.¹⁹ After completion of CDC-recommended isolation,⁶ patients were assessed in clinic with chest X-ray imaging, formal spirometry, and measurement of donor-specific anti-HLA antibodies using a solid-phase flow cytometry assay (Luminex)²⁰. Bronchoscopy with bronchoalveolar lavage and transbronchial biopsies was performed if there was a >10% decline in FEV₁ compared to pre-COVID baseline values or if there were chest imaging abnormalities that persisted beyond the first 4–6 weeks, unless contraindicated.²¹ Tissue specimens were assessed by two pulmonary pathologists and graded for ACR and other pathology according to ISHLT guidelines.²² Persistent non-infectious ALAD was treated with either corticosteroids (methylprednisolone 7.5mg/kg IV daily x 3 days, followed by prednisone taper), anti-thymocyte globulin, or plasmapheresis (PLEX) with IVIg.²³ Probable antibody mediated rejection was treated with PLEX, IVIg, and rituximab.³⁵

Primary and secondary clinical outcomes

The primary study endpoint was a change in rate of FEV₁ before and after SARS-CoV-2 infection. Patients with at least two FEV₁ measurements (separated by 30 days) before and after COVID infection were included, and the change in FEV₁ or FVC (mL/day) was calculated by identifying the slope of trendlines pre-COVID and post-COVID. Spirometry was performed at the VUMC Pulmonary Function Lab according to ATS/ERS guidelines.³⁶ Secondary outcomes included incidence of ALAD, AMR, dnDSA, hospitalization, intubation, length of hospital stay, secondary infections, and death.³⁵

Statistics

Continuous variables were described as median with interquartile range and categorial variables as proportions. The rates of change in FEV₁ and FVC (mL/day) before and after SARS-CoV-2 infection were compared using Wilcoxon ranked sum testing. Comparisons between SARS-CoV-2 eras (pre-Delta, Delta, or Omicron) were performed using Kruskal-Wallis testing for continuous variables and Chi-squared testing for categorical variables. Univariate analysis to assess for ALAD risk factors was performed using a Fisher’s exact test for categorical variables and Mann-Whitney U tests for continuous variables.

RESULTS

Study population

A total of 103 patients developed SARS-CoV-2 during the study period (Table 1), representing 30.7% of the VUMC living lung transplant recipient cohort (n=336). Of these, 27 patients were classified as having pre-Delta-era infections, 20 patients with Delta-era infections, and 56 patients with Omicron-era infections. The median follow- up time after SARS-CoV-2 diagnosis was 123 days (IQR 98–213). Patients acquired COVID-19 a median of 3.5 years post-transplant (IQR 1.6 – 6.5). Most patients (91%) had upper or lower respiratory tract infection symptoms. The majority of patients (77%) were CLAD-free when infected with COVID, while 14% had CLAD stage 1 and 8% of patients had CLAD stage 2 or greater. Patient demographics and maintenance immunosuppression were similar across SARS-CoV-2 eras (Table 1). Most patients (69%) received outpatient monoclonal antibody therapy, with increased administration to patients during the Delta and Omicron eras. In our cohort, 25 patients (24%) required hospitalization for acute SARS-CoV-2 infection, with 7 (7%) patients requiring mechanical ventilation (invasive or non-invasive). A total of 10 patients (10%) died, 6 of whom died during the initial hospitalization for acute COVID-19. Rates of hospitalization, intubation, and length of stay did not differ between SARS-CoV-2 variants. For patients in the Delta variant era, a greater concentration of supplemental oxygen was required.

Table 1:

Baseline characteristics of the study cohort

	Total SARS-CoV-2 (N=103)	SARS-CoV-2 Pre-Delta Era (N=27)	SARS-CoV-2 Delta Era (N=20)	SARS-CoV-2 Omicron Era (N=56)	p value
Age (years)	62 (52–68)	58 (47–64)	62 (55–68)	63 (46–68)	0.15
Female Sex	49 (48%)	13 (48%)	10 (50%)	26 (46%)	0.96
White Race	87 (84%)	24 (89%)	17 (85%)	46 (82%)	0.73
LAS at Transplant	39.5 (35.9–49)	40.4 (35.2–49.8)	39.4 (34.4–46.7)	39.9 (37.1–48.0)	0.75
Bilateral Transplant	80 (78%)	21 (78%)	14 (70%)	45 (80%)	0.63
Transplant Indication					0.49
Obstructive	20 (19%)	5 (19%)	5 (25%)	10 (18%)
Pulmonary Vascular	3 (3%)	1 (4%)	0 (0%)	2 (4%)
CF/ Bronchiectasis	24 (23%)	10 (37%)	2 (10%)	12 (21%)
Diffuse Parenchymal	56 (54%)	11 (41%)	13 (65%)	32 (57%)
Time from transplant to SARS-CoV-2 (years)	3.5 (1.6–6.5)	4.4 (2.6–7.1)	3.0 (1.9–5.4)	3.3 (1.6–6.5)	0.34
BMI*	27.3 (23.7–30)	27.3 (23.1–29.5)	27.8 (25.4–29.5)	26.5 (23.8–30.3)	0.80
Diabetes*	37 (36%)	13 (48%)	6 (30%)	18 (32%)	0.30
Calcineurin Inhibitor Type*					0.76
Tacrolimus	94 (91%)	24 (89%)	19 (95%)	51 (91%)
Cyclosporine A	9 (9%)	3 (11%)	1 (5%)	5 (9%)
Anti-proliferative Type*					0.44
Mycophenolate	54 (52%)	15 (56%)	8 (40%)	31 (55%)
Azathioprine	13 (13%)	4 (15%)	1 (5%)	8 (14%)
Sirolimus	5 (5%)	2 (7%)	1 (5%)	2 (4%)
No Anti-proliferative	31 (30%)	6 (22%)	10 (50%)	15 (27%)
Prednisone	103 (100%)	27 (100%)	20 (100%)	56 (100%)	1.00
CLAD Stage*					0.43
0	79 (77%)	23 (85%)	18 (90%)	40 (71%)
1	14 (14%)	1 (4%)	1 (5%)	11 (20%)
2	3 (3%)	2 (7%)	0 (0%)	1 (2%)
3 or 4	6 (6%)	1 (4%)	1 (5%)	4 (7%)
Symptomatic SARS-CoV-2	94 (91%)	24 (89%)	18 (90%)	52 (93%)	0.81
SARS-CoV-2 Vaccinated*	69 (67%)	0	17 (85%)	53 (95%)	<0.01
mRNA vaccine (2 doses)		0	15 (88%)	51 (96%)	0.25
Non-mRNA vaccine		0	2 (12%)	2 (4%)
Time from Vaccine to COVID	269 (175–298)	n/a	156 (126–176)	281 (261–310)	<0.01
Tixagevimab / Cilgavimab	4 (4%)	0	0	4 (7%)	0.89
Monoclonal Antibody Therapy	71 (69%)	13 (48%)	17 (85%)	41 (73%)	0.02
Other Therapies	22 (21%)	5 (19%)	6 (30%)	11 (20%)	0.57
Remdesevir	15 15%)	2 (7%)	5 (25%)	8 (14%)
Dexamethasone / Steroids	19 (18%)	3 (11%)	6 (30%)	10 (18%)
Baricitinib	3 (3%)	0	2 (10%)	1 (2%)
Molnupiravir	1 (1%)	0	0	1 (2%)
Time first follow-up appt (d)	30 (27–39)	27 (25–37)	33 (28–40)	32 (28–38)	0.15
Days follow-up after COVID	123 (98 – 213)	414 (376 – 439)	147 (133 – 163)	100 (72–114)	<0.01

^*At time of SARS-CoV-2 diagnosis

Table indicates n (%) for categorical variables or median (interquartile range) for continuous variables.

SARS-CoV-2 infection did not change the rate of decline of FEV₁ or FVC in lung transplant recipients

A minimum of two spirometric assessments (separated by at least 30 days) both pre-COVID and post-COVID were available for 85 patients (23 patients pre-Delta-era, 18 patients in the Delta-era, and 44 patients in the Omicron-era). There were 18 patients who did not meet the criteria for inclusion in spirometry analysis (6 died during the index hospital stay, 2 discharged to hospice, 1 unable to perform spirometry due to pre-existing chronic comorbidity, 1 without post-transplant FEV₁ baseline, and 8 who did not have two spirometric assessments 30 days apart). Overall, the rate of FEV₁ and FVC change did not differ pre- and post- COVID-19 disease. The median rate of FEV₁ change pre-COVID was −0.05 mL/day (IQR −0.50 – 0.60) compared to −0.20mL/day (IQR −1.40 – 0.70) post-COVID (p= 0.16) (Figure 1, Table 2). The median rate of change in FVC pre-COVID was 0.20 mL/day (IQR −0.60 – 0.70) compared to 0.05mL/day (IQR −1.00 – 1.10) post-COVID (p= 0.76). During the Delta era, there was a significant difference in FEV₁ change pre-COVID compared to post-COVID [median FEV₁ change pre-COVID 0.25 mL/day (IQR −0.06 – 0.60) compared to −0.20 mL/day (IQR −0.58 – 0.15 mL/day) post-COVID, p=0.03]; in contrast, there was no significant change in rate of FEV₁ decline in the pre-Delta or Omicron eras. There was no significant impact of SARS-CoV-2 variant on FVC change over time.

Figure 1. The effect of SARS-CoV-2 infection on rate of change of FEV₁ in lung transplant recipients.

FEV₁ rate of change was calculated before and after SARS-CoV-2 infection for 85 lung transplant recipients with two spirometric measurements at least 30 days apart after completion of isolation after acute SARS-CoV-2 infection. The slope of FEV1 change over time was calculated over 3–6 months before SARS-CoV-2 infection and for a minimum of 3 months after SARS-CoV-2 infection. Change in the slope of FEV1 decline before and after SARS-CoV-2 were compared by Wilcoxon ranked sum testing. A) Pre-Delta era, n=23. B) Delta era, n=18. C) Omicron era, n=44.

Table 2:

Clinical Outcomes in Lung allograft Recipients with COVID

	Total SARS-CoV-2 (N=103)	SARS-CoV-2 Pre-Delta Era (N=27)	SARS-CoV-2 Delta Era (N=20)	SARS-CoV-2 Omicron Era (N=56)	p value
Hospitalized for SARS-CoV-2	25 (24%)	8 (30%)	7 (35%)	10 (18%)	0.23
Peak FiO2	0.30 (0.21 – 1.0)	0.21 (0.21 – 0.24)	0.66 (0.27 – 1.0)	0.35 (0.23 – 0.93)	0.39
Mechanical Ventilation	7 (7%)	1 (4%)	3 (15%)	3 (5%)	0.26
Hospital Length of Stay (d)	6 (5–11)	5 (4 – 7)	5 (5 – 14)	8 (6 – 17)	0.33
Death within 6 months of COVID	10 (10%)	3 (4%)	3 (6%)	4 (7%)	0.57
In-hospital / acute COVID	6 (6%)	1 (4%)	2 (10%)	3 (5%)	0.64
Post-acute infection	4 (4%)	2 (7%)	1 (5%)	1 (2%)	0.44
CXR changes >4 wk post-COVID diagnosis	15 (15%)	6 (22%)	5 (25%)	4 (7%)	0.06
Accelerated CLAD or death In patients with prior CLAD	11/22 (50%)	3/4 (75%)	1/3 (33%)	7/15 (47%)	0.50
Probable AMR (< 6 mo post-COVID)	3 (3%)	0	1 (5%)	2 (4%)	0.98
Secondary infections within 90 d*	14 (14%)	1 (4%)	2 (10%)	11 (20%)	0.12
Cytomegalovirus viremia	4 (4%)	0	0	4 (7%)
Community acquired viral infection	5 (5%)	0	1 (5%)	4 (7%)
Bacterial pneumonia	7 (7%)	0	1 (5%)	5 (9%)
Staphylococcus species	3 (3%)	0	0	2 (4%)
Gram negative rods	4 (4%)	0	1 (5%)	4 (8%)
Bacteremia (MRSA)	0	1 (4%)	0	0
	Total SARS-CoV-2 (N=85)	SARS-CoV-2 Pre-Delta Era (N=23)	SARS-CoV-2 Delta Era (N=18)	SARS-CoV-2 Omicron Era (N=44)	p value
Change in rate of FEV1 decline before and after COVID (mL/d)	−0.60 (−1.73 – 0.50)	0.16 (−1.11 – 1.10)	−0.69 (−1.35 to −0.15)	−1.05 (−2.53 – 0.40)	0.19
Change in rate of FVC decline before and after COVID (mL/d)	−0.20 (−1.50 – 1.10)	0.60 (−0.45 – 1.20)	−0.54 (−0.80 – 0.33)	−0.75 (−2.63 – 1.38)	0.07
Acute decline in FEV1 >10% baseline values (ALAD)	26 (31%)	7 (30%)	5 (28%)	14 (32%)	0.95
For-cause bronchoscopy (ALAD)	16 /26 (62%)	5/7 (71%)	3/5 (60%)	8/14 (57%)	0.31
Etiology of ALAD
ACR	3 (4%)	1 (4%)	0	2 (5%)
Organizing pneumonia	4 (5%)	0	2 (11%)	2 (5%)
Super-imposed infection	1 (1%)	0	1 (6%)	0
Inflammation, non-specific	8 (9%)	4 (17%)	0	4 (9%)
Treatment of Post-COVID ALAD#	9 /26 (35%)	3 /7 (43%)	2/5 (40%)	4/14 (29%)	0.60
Steroids	9 (35%)	3 (43%)	2 (40%)	4 (29%)
Anti-thymocyte globulin	1 (4%)	0	0	1 (7%)
Plasma Exchange and IVIg	2 (8%)	0	0	2 (14%)
ALAD Recovery	9/26 (35%)	3/7 (43%)	2/5 (40%)	4/14 (29%)	0.59
Recovery following treatment	5/10 (50%)	2/3 (66%)	1/3 (33%)	2/4 (50%)	0.72
Recovery without treatment	4/16 (25%)	1/4 (25%)	1/2 (50%)	2/10 (20%)	0.67

Table indicates n (%) for categorical variables or median (interquartile range) for continuous variables.

^*Some patients had multiple infections

^#Some patients had combination therapies

Clinical course after SARS-CoV-2

Among the cohort of COVID-19 survivors (both hospitalized and non-hospitalized) that performed follow-up lung function testing (N=85), 33 patients developed ALAD or accelerated CLAD, defined as FEV₁ decline >10% of baseline pre-COVID values. Of these patients, 26 patients (31%) met criteria for ALAD, while the remaining 7 had accelerated CLAD. A total of 19 patients (16 with ALAD and 3 with accelerated CLAD) underwent diagnostic bronchoscopy (Table 2). Common pathologic findings for the patients with ALAD included non-specific inflammation (11 patients), organizing pneumonia (5 patients), and biopsy-proven ACR (3 patients). For the patients with accelerated CLAD, diagnostic bronchoscopy revealed non-specific inflammation in 2 patients and was non-diagnostic in the other. Nine of the 26 patients (35%) with ALAD underwent treatment with either corticosteroids, anti-thymocyte globulin, or PLEX with IVIg. After 3 months of follow up, 9/26 (35%) patients with ALAD had recovery of FEV₁ to within 10% of their pre-COVID values. There was a trend towards improved FEV₁ in patients that received treatment for ALAD, though this was not statistically significant (50% vs 25%, p= 0.23). Twenty-two patients in the cohort met criteria for CLAD prior to acquiring COVID-19. Four (18%) of those patients died. As described above, 7 patients with CLAD experienced a >10% decline in FEV₁ compared to pre-COVID values within 3 months of contracting SARS-CoV-2. A univariate analysis was performed to assess for risk factors for ALAD or accelerated CLAD after SARS-CoV-2. Lung allocation score was associated with increased ALAD risk, whereas azathioprine was associated with reduced ALAD risk. Diabetes was associated with increased risk for ALAD, but not progressive CLAD. Pre-existing CLAD, non-vaccinated status, and diffuse parenchymal lung disease as the underlying indication for transplant did not reach statistical significance as risk factors for ALAD (See Table 3). Secondary infection occurred in 14% of the cohort and was not associated with SARS-CoV-2 variant era type. Bacterial pneumonia was the most common infection, followed by community acquired viruses and CMV viremia.

Table 3:

Univariate Analysis Risk Factors for Post-COVID ALAD or Accelerated CLAD^*

	ALAD or Accelerated CLAD (N= 33)	No Allograft Dysfunction (N=52)	P value
ALAD	26 (79%)	0
Accelerated CLAD	7 (21%)	0
COVID Variant Era
Pre-delta	10 (30%)	13 (25%)	0.62
Delta	5 (15%)	13 (25%)	0.41
Omicron	18 (55%)	26 (50%)	0.82
Symptomatic COVID	32 (97%)	46 (88%)	0.24
Need for Hospitalization	8 (24%)	8 (15%)	0.40
COVID Vaccination	18 (55%)	38 (73%)	0.10
Receipt of Monoclonal Ab	22 (66%)	38 (73%)	0.63
Time from Transplant to COVID Diagnosis (years)	3.49 (2.12 – 6.60)	3.39 (1.57 – 6.40)	0.75
Age	62 (52 – 67)	62 (54 – 67)	0.99
Female sex	15 (45%)	28 (54%)	0.51
White Race	27 (82%)	45 (87%)	0.55
Bilateral Lung Transplant	25 (76%)	44 (85%)	0.40
Lung Allocation Score	42.24 (38.36 – 52.02)	38.71 (35.23 – 46.41)	0.03
Indication for Lung Transplant
Obstructive	2 (6%)	13 (25%)	0.04
Pulmonary Vascular	0	3 (6%)	0.28
CF/ Bronchiectasis	9 (27%)	11 (21%)	0.60
Diffuse Parenchymal	22 (66%)	25 (48%)	0.12
Pre-existing CLAD	11 (33%)	8 (15%)	0.07
Diabetes	13 (39%)	16 (31%)	0.48
Diabetes as a risk factor for ALAD only	12/26 (46%)	16/59 (27%)	0.13
Last known estimated glomerular filtration rate pre-COVID (mL/min)	43 (29 – 56)	43 (30 – 56)	0.85
Body Mass Index	26.7 (23.6 – 29.9)	27.8 (24.3 – 30.1)	0.79
Antiproliferative Type
Mycophenolate	19 (58%)	24 (46%)	0.38
Azathioprine	1 (3%)	10 (19%)	0.04
Sirolimus	2 (6%)	3 (6%)	1.0
None	11 (33%)	15 (29%)	0.81

^*For patients able to complete >2 spirometry assessments (>30 days apart) pre- and post-COVID

Impact of Vaccination Status on Clinical Outcomes

Because of the timing of vaccine availability, none of the lung transplant recipients that acquired pre-Delta SARS-CoV-2 variants were vaccinated. Two doses of mRNA vaccine were completed in 15 patients in the Delta era and 51 patients in the Omicron era, with 2 additional patients in each group completing non-mRNA vaccination (Table 1). The median time between completion of the primary vaccination series and SARS-CoV-2 diagnosis was 269 days (156 days for Delta, 281 for Omicron). Vaccination status had no significant impact on FEV₁ trajectory pre- and post-COVID infection; the median change in FEV₁ in vaccinated patients was −0.85 mL/day (IQR −2.14 to 0.25) compared to 0 mL/day (IQR −1.42 to +1.10) in unvaccinated patients. There was an association between vaccination status and change in FVC [−0.71 mL/day (IQR −2.30 to −.90) in vaccinated patients vs. 0.50 ml/day (IQR −0.70 to 1.10) in unvaccinated patients, p= 0.02]. Vaccination status had no detectable impact on rates of hospitalization, intubation, or death from acute COVID-19 (Supplemental Table 1, p>0.1 for each). Vaccination status was not associated with AMR or acceleration of CLAD. There was a non-significant trend towards higher incidence of persistent chest radiograph abnormalities and ALAD in vaccinated patients (p=0.07 and 0.14, respectively).

DISCUSSION

In this single-center, observational study, we report the clinical course of lung transplant recipients infected by SARS-CoV-2 from initial illness through 3 months of follow-up. Overall, SARS-CoV-2 infection was not associated with a change in the rate of change of FEV₁ or FVC. However, 39% of lung transplant recipients developed ALAD or accelerated CLAD after SARS-CoV-2 infection, with approximately one-third recovering lung function back to baseline during the follow up period. Non-specific inflammation was the most common pathology found on lung biopsy during recovery. We identified potential risk factors for ALAD including higher LAS at the time of transplant, lack of azathioprine use, and diabetes.

Our findings build on a small cohort of patients described by Permpalung et al., who had an acute FEV₁ decline of >10% (ALAD) in 20% of their cohort. Rates of biopsy-proven ACR, dnDSA, and AMR were similarly low¹⁵. Mahan et al. found a higher rate of ALAD in their cohort, with almost half of patients having a sustained decline in FEV₁ or FVC >10%.¹⁴ This may be due to higher rates of hospitalization (88%), lower rates of monoclonal antibody infusion (13.6%), or infection with pre-Delta variants. Our data also build on these previous studies by specifically assessing for differential impact of SARS-CoV-2 variants on clinical outcomes. We showed that infection with the Delta variant was associated with a change in the rate of FEV₁ decline, but we did not find spirometric differences in pre-Delta or Omicron-era infections. We found that the rate of secondary bacterial infections was greatest during the Omicron era, despite an overwhelming majority of patients being vaccinated and receiving appropriate monoclonal antibody therapy within the first few days of their illness.

Seventy three percent of the patients that developed ALAD did not require hospitalization for their acute SARS-CoV-2 infection, suggesting that minimal disease from SARS-CoV-2 is sufficient to trigger allograft injury. Our center chose to treat patients surviving COVD-19 who had persistent lung dysfunction for presumed immune-mediated lung injury with augmentation of immunosuppression with steroids, ATG, or PLEX with IVIg, depending on the clinical context. Augmented immunosuppression for COVID-related ALAD was associated with increased recovery of FEV₁, suggesting that COVID-induced lung inflammation in lung transplant recipients may be potentially reversible. In addition, we identified a trend that patients with diffuse parenchyma lung disease may have increased ALAD risk. Others have previously reported an association between telomere length and COVID-19 severity.^37,38 One possible explanation of these findings could be that genetic telomeropathies in the setting of diffuse parenchymal lung disease may alter immune responses during post-COVID ALAD. The potential protective benefit of azathioprine is also interesting. Multiple groups have identified a decreased humoral immune response with mycophenolate in the setting of vaccine responsiveness.³⁹ It is possible that the antiproliferative immunosuppressants may potentiate early pro-inflammatory responses following infection with SARS-CoV-2. Further mechanistic work is needed to define the specific inflammatory pathways triggered by SARS-CoV-2 and whether targeted therapy is possible in the future to reduce the risk of ALAD. The finding that SARS-CoV-2 pneumonia can precipitate persistent inflammation may be similar to allograft injury triggered by other community-acquired respiratory viral infections, which is well-described.^25–28 Longer follow-up of this patient cohort is needed to determine whether SARS-CoV-2 infection increases risk of CLAD.

The rate of infection in vaccinated individuals in our cohort (16/1000 patient months) is similar to other reports in solid organ transplant recipients²⁹. Our lack of mortality benefit from vaccination differs from one previous registry study in the United Kingdom that showed reduced mortality from COVID-19 in solid-organ transplant recipients who were fully vaccinated,¹³ though the number of lung transplant recipients included in the registry was not stated. Alternatively, differences in compliance with public health measures may differ between the UK and the southeastern United States. Our study suggests that vaccinated lung transplant patients remain at risk for SARS-CoV-2-induced ALAD, a finding that will require additional follow-up in larger study cohorts. We were unable to statistically test the impact of supplemental booster doses of vaccination in our cohort, because the majority of Delta infections occurred prior to booster dose administration and the majority of Omicron infections occurred despite booster dosing. Additional doses of mRNA vaccines and prophylactic monoclonal antibodies like tixagevimab/cilgavimab may provide additional protection from poor clinical outcomes after SARS-CoV-2.

Our study has several strengths. This is a large cohort study specifically focused on lung transplant recipients in a real-world management setting. Comparison of changes in FEV₁ over time pre-and post-COVID infection is more robust than comparison of singular spirometric values. Changes in spirometry provide more significant prognostic value than absolute values.³⁰ In addition, use of each patient’s previous spirometry over time strengthened the statistical power of our analysis. We show that the impact of SARS-CoV-2 changes with different variants and that increased immunosuppression after initial infection may facilitate lung function recovery. Our study also has some important limitations. This is a retrospective single-center study. Diagnostic assays for COVID varied widely from hospital-based PCR tests to home kits, raising the possibility that some false-positive tests were included; since 91% of patients were symptomatic at the time of COVID diagnosis, the false positive rate was likely low. We did not collect data on specific symptoms at the time of COVID diagnosis to allow for delineation of upper versus lower respiratory tract disease. Several patients were unable to complete the required spirometric follow up to be included in the analysis, potentially introducing selection bias in our conclusions. Differences in local public health compliance and vaccination adherence may impact the generalizability of our findings. None of our patients with pre-Delta infections had been vaccinated prior to SARS-CoV-2 diagnosis, so we were unable to assess vaccine efficacy in this group.

In conclusion, in an observational cohort of lung transplant recipients, we demonstrate that SARS-CoV-2 infection has limited impact on trajectory of lung function. While many survivors developed transient ALAD, augmentation of immunosuppression promoted recovery of lung function in a minority of patients. This study provides key information about short-term clinical outcomes after COVID infection in lung transplant recipients and provides data that can guide design of future interventional studies to improve clinical outcomes.

Abbreviations:

ACR

acute cellular rejection

AFOP

acute fibrinous organizing pneumonia

AMR

antibody-mediated rejection

ALAD

acute lung allograft dysfunction

CARV

Community-acquired respiratory viruses

CLAD

chronic lung allograft dysfunction

CMV

cytomegalovirus

COVID

coronavirus disease

dnDSA

de novo donor specific antibodies

FEV₁

Forced Expiratory Volume in 1 second

FiO₂

fraction of inspired oxygen

FVC

Forced Vital Capacity

ISHLT

International Society of Heart and Lung Transplantation

IVIg

intravenous immunoglobulin

LAS

Lung allocation score

LOS

length of stay

OP

organizing pneumonia

PLEX

Plasma exchange

SARS-CoV-2

severe acute respiratory syndrome coronavirus 2

SOT

solid organ transplant

VUMC

Vanderbilt University Medical Center

Data Sharing and Data Accessibility

The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary materials. Any other reasonable information is available from the corresponding author (AJT) upon request.