Progress and treatment of “long COVID” in non‐hospitalized patients: A single‐center retrospective cohort study

“Long COVID” or “post‐COVID conditions” describes prolonged symptoms after the acute phase of coronavirus disease 2019 (COVID‐19). However, there is a paucity of published reports on its treatment.


INTRODUCTION
Coronavirus disease 2019 , caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has a staggering health impact worldwide with many infections and deaths. 1 COVID-19 not only progresses to severe illness but also results in prolonged symptoms after the acute phase, known as "long COVID" or "post-COVID conditions" 2 (hereafter referred to as long COVID). Other viruses can also cause chronic respiratory symptoms due to interstitial changes in the lungs 3 and disturbances in the neural pathways associated with coughing. 4 However, long COVID is characterized by the persistence and development of various systemic symptoms other than respiratory symptoms including physical, cognitive, and psychological symptoms. [5][6][7] Long COVID results in reduced quality of life (QoL) 8,9 ; however, its effective management is unknown, other than rehabilitation. 10,11 We postulated that it would be beneficial to collect and evaluate reports on the course of long COVID and its treatment to formulate a more precise treatment strategy in the future.
We have used Japanese traditional medicine, i.e., Kampo medicine, to treat patients with COVID-19 not only in the acute phase but also in the chronic phase. Kampo medicine sometimes treats symptoms that are untreatable with Western medicines, such as fatigue. It is characterized by its ability to treat various symptoms with a single prescription including multiple bioactive components from crude drugs. We have previously shown that treatment including Kampo could promote the improvement of olfactory dysfunction in patients with COVID-19. 12 In this report, we describe the progress of non-hospitalized, adult patients with long COVID whose treatment included Kampo medicine in a cohort study.

METHODS
In Japan, in the early phase of the pandemic, patients with confirmed COVID-19 who were asymptomatic or had mild symptoms were observed at isolation facilities. In Miyagi prefecture, medical doctors in the Department of General Medicine and Kampo Medicine, Tohoku University Hospital visited the isolation facilities to check the condition of patients with poor vital signs or moderate symptoms and performed triage to decide whether to hospitalize or treat them in the isolation facilities. 13 Patients who underwent medical evaluation were checked by telephone one and two weeks after the first examination; if they had prolonged symptoms, medical examination was continued via telephone or at the hospital outpatient setting. The prescriptions were decided by the physicians according to the clinical course, physical signs, symptoms, and laboratory findings relating to each patient.
This study was a retrospective, single-center, cohort study. The inclusion criteria were adult COVID-19 patients who underwent medical examinations at the isolation facilities in the Miyagi prefecture and had COVID-19-related symptoms at least one month after disease onset, 14 between October 2020 and September 2021. The observational period was between October 2020 and March 2022. The exclusion criteria were age less than 20 years old and hospitalized after our medical examination.
The primary endpoints were symptoms resolution and "residual ratio" of each symptom. "Symptom resolution" was defined as the disappearance of symptoms or the presence of mild symptoms not requiring treatment. "Residual ratio" was calculated as the number of symptomatic patients at six months divided by the number of symptomatic patients at one month. Patients' characteristics, symptoms, and medications prescribed were extracted from the medical records of the hospital. Symptoms and medications were extracted for four periods: 1-2 months; 2-3 months; 3-6 months; and over 6 months from the onset of COVID-19. The severity of COVID-19 was classified according to the Clinical Practice Guidance Review Committee 2022 15 : patients with a SpO 2 of 96% or higher in the absence of imaging evaluation were classified as having "mild"; patients with a SpO 2 of less than 96% or pneumonia on imaging were classified as having "moderate I"; patients with a SpO 2 of no more than 93% and requiring oxygen were classified as having "moderate II"; and patients who required artificial ventilation or needed to stay in the intensive care unit were classified as having "severe" COVID-19. All the Kampo formulae used in this report were manufactured by Tsumura & Co. (Tokyo, Japan). The detailed drug information of Kampo formulae is listed in STORK (http://mpdb. nibiohn,go.jp/stork/).

RESULTS
During the study period, 1146 patients underwent examinations, and of these, 70 (6.1%) eligible patients (females: 37 [52.9%]) were included ( Figure 1). The characteristics of the patients are detailed in Table 1 complete symptom resolution during the observational time, and 32 patients (46.5%) had their followup visits interrupted. Twenty-nine patients (41.4%) were examined at least two months from the disease onset, 31 patients (30.0%) at least three months, and nine patients (12.9%) after six months or more.
Of the nine patients examined over six months from disease onset, five recovered completely within the study period, whereas four patients (5.7%) had residual symptoms. In addition, seven of nine patients were women (77.8%), with a median age of 45 years (range: 43-55 years). Regarding severity, four of the nine patients (44.4%) had "mild" and five (63.6%) had "moderate I" disease.
The symptoms at one month after the first examination and residual ratios are presented in Table 2. The patients had multiple symptoms, and the median number of symptoms was three (range: 1-10) at one month. The most common symptoms at one month were cough (n = 33, 47.1%), followed by olfactory dysfunction (n = 27, 38.6%), fatigue (n = 21, 30.0%), and taste dysfunction (n = 17, 24.3%). Other symptoms that persisted at one month were itchiness, numbness, stiff shoulders, hoarseness of voice, oral and tongue discomfort, chills, constipation, and varicose veins of the lower extremities.
Cough, the most frequent symptom at one month, had a low residual rate of 3.0%, followed by less frequent symptoms like olfactory dysfunction, fatigue, and taste dysfunction, all of which had relatively high residual rates of 7.4%, 14.3%, and 11.8% respectively. The other symptoms with high residual rates were dizziness (33.3%), musculoskeletal pain (14.3%), and abdominal discomfort (14.3%).

DISCUSSION
We showed the clinical course of long COVID with treatment including Kampo medicine in nonhospitalized adult patients. Many previous studies are epidemiological reports about the frequency of symptoms after a certain period of time from the onset of COVID-19. Regarding the treatment of long COVID, rehabilitation and physical exercise are reported as potentially effective. 10,33 However, effective pharmacological treatment of long COVID is largely unknown.
Non-hospitalized patients have reported symptom persistence rates of 53.1% at an average of 125 days, 19 64.2% at 3-4 months, 20 and 32.7% at six months. 21 Recently, in Japan, 32.3% of the patient group, including 86.4% with mild disease in the acute phase had persistent symptoms at six months. 22 In this study, the rate of patients who need medical examination was 13% at six months. Our residual-symptom data cannot be simply compared with previous epidemiologic persistence rates, because the study design is different. However, this study provided important findings that the treatment involving Kampo medication may alleviate patients suffering from long COVID.
Other common viruses can cause various postinfectious symptoms: cough, 4 fatigue, 17 olfactory dysfunction, and taste dysfunction. 18 Long COVID may be gaining medical attention because of the numerous patients with COVID-19 leading to a large number of patients with long COVID.
The symptoms may occur or be aggravated due to social factors, such as isolation from society. COVID-19 T A B L E 3 Medications prescribed for each duration from the onset of COVID-19.  [Correction added on 13 June 2023, after first online publication: Table 3 has been updated to include relevant subheadings in the 'Kampo medication' section and the layout has been slightly adjusted].
was reported to have negative psychological effects, 23 and mental health and cognitive function decreased under restriction or lockdown. 24 However, long COVID is considered more severe and frequent than other common post-viral infectious symptoms. For example, the prevalence of olfactory dysfunction in COVID-19 is reportedly nearly three times as high as in other viral infections. 25 SARS-CoV-2 uses human angiotensin-converting enzyme 2 (ACE2) receptors for host-cell entry 26 and binds more effectively to ACE2, compared with SARS-CoV. 27 ACE2 is distributed throughout the body; hence, COVID-19 causes a systemic inflammatory response and dysfunction in various organs 28 ; the inflammatory response persists for a long time. 29 In addition, vascular endothelium expresses ACE2, and its damage and inflammation interfere with oxygen supply to the tissue and exacerbates inflammation and tissue damage. This is an important factor in the mechanism of long COVID. 30 It has also been argued that long COVID may be due in part to persistent viral infection. 31 In this study, females aged 40-50 years tended to have prolonged symptoms. The female sex has been reported as a risk factor for long COVID. 32 Women had exacerbated anxiety and depression under the COVID-19 lockdown and were more affected by home confinement than men. 23,33 Moreover, ACE2 is also present in the reproductive organs 28 and may have a stronger effect in perimenopausal women, who are naturally prone to having several complaints. Some of their complaints due to long COVID symptoms may overlap with menopausal symptoms.
Long COVID should be treated considering the pathogenesis and the individual variations, such as original constitution, comorbidities, and social environment. Hence, a holistic and comprehensive medical approach is important. In Kampo medicine, comprehensive physical findings result in multiple symptoms being converted to simple diagnoses and prescriptions. This traditional theory may be appropriate in the treatment of long COVID. In the previous report on the clinical course of long COVID, many Kampo formulae were also used. 16 On the other hand, the fact that many kinds of formulae were used indicates that treatment is difficult with specific prescriptions. It was also considered necessary to make changes depending on the term from onset.
In the present study, Kampo formulae with antiinflammatory effects that were frequently used in the early stage included shosaikoto, formulae involving shosaikoto, 34,35 keigairengyoto, 12 and daisaikoto. Some crude drugs of Kampo formulae, for example, glycyrrhizae radix, 36 a crude drug frequently included in Kampo formulae, are reported to have anti-SARS-CoV-2 activity and an anti-inflammatory effect. Considering the pathology that persistent inflammation or persistent infection are related to the onset of long COVID, such properties may be advantageous in treating the subsequent progression of long COVID.
Tonifying formulae, such as ninjin'yoeito, hochuekkito, and juzentaihoto, which assist in general recovery from illness, were frequently used throughout the entire period. For example, hochuekkito has been reported to have immunomodulatory effects pharmacologically, and clinically improve fatigue and weakness. 37 Previously, hochuekkito was reported to be the most commonly used formula for the treatment of general fatigue in long COVID. 38 Thus, tonifying formulae are considered to be a useful treatment option for fatigue in long COVID.
In this study, the frequency of prescribing blood stasis-resolving formulae such as keishibukuryogan, kamishoyosan, and tokishakuyakusan increased in the late stage of long COVID. These formulae are frequently used for treating indefinite menopausal symptoms and chronic pain. Chronic symptoms activate the sympathetic nervous system and increase peripheral vascular resistance, which is often present in patients with blood stasis conditions. For example, keishibukuryogan was reported to have protective effects on the endothelial function and treat neurotoxicity in basic research. 39 Paeonol in Paeonia suffruticosa, one of the crude drugs used to resolve blood stasis, improves endothelium injury. 40 Endothelial damage is considered responsible for long COVID; therefore, blood stasisresolving formulae may be suitable for treatment.
The present study had certain limitations including several factors related to the treatment prognosis. Since this is a retrospective cohort study, we could not present exact residual-symptom rates because of possible interruptions during the observation period. The percentage of unvaccinated patients was high because the vaccine was not widely available during the study period. The treatments were prescribed by four Kampo specialists and three Kampo training doctors; hence their diagnostic and treatment skills varied. The treatment effect was not confirmed, because the patients may have spontaneously recovered. However, clinically it is difficult to make the control group with no medication. The Japan Society for Oriental Medicine has planned clinical trials and research of Kampo medicine in patients with COVID-19-related sequelae, 41 so future results are awaited.

CONCLUSION
Non-hospitalized patients with COVID-19 may suffer from multiple persistent symptoms after the acute phase of infection. For the management of long COVID, a comprehensive and holistic approach is needed. Kampo medicine should be considered as a treatment option for long COVID.
CONFLICT OF INTEREST STATEMENT Ryutaro Arita, Akiko Kikuchi, Minoru Ohsawa, Shin Takayama, and Tadashi Ishii are associated with the Department of Kampo and Integrative Medicine, Tohoku University School of Medicine. The department received a grant from Tsumura & Co, a Japanese manufacturer of Kampo medicine; however, the grant was used per Tohoku University guidelines. Potential conflicts of interest were addressed by the Tohoku University Benefit Reciprocity Committee and were managed appropriately. There are no other competing interests to declare.

APPROVAL OF THE RESEARCH PROTOCOL BY AN INSTITUTIONAL REVIEWER BOARD AND THE APPROVAL NUMBER
The observational study was approved by the ethical committee of Tohoku University (Miyagi, Japan) (approval number: 2021-1-447). The case reports were approved by the ethical committee of Tohoku University (Miyagi, Japan) (approval number: 28256).