Implications of COVID-19 in high burden HIV/TB countries: A systematic review of evidence

The triple burden of COVID-19, tuberculosis and human immunodeciency virus is one of the major global health challenges of the 21st century and in the future. In high burden HIV/TB countries, the spread of COVID-19 among people living with HIV is a well-founded concern. A thorough understanding of HIV/TB and COVID-19 pandemics is important as the three diseases interact. This may clarify HIV/TB/COVID-19 as a newly related eld and play an important role in the present and future management of the co-infections. However, several gaps are remaining in the knowledge of the burden of COVID-19 on patients with TB and HIV, the diagnosis, and management of these patients. The study was conducted to review different studies on SARS-CoV, MERS-CoV or COVID-19 associated with HIV/TB coinfection or only TB and to understand the interactions between HIV, TB and COVID-19 and its implications on the burden of the COVID-19 among HIV/TB co-infected or TB patients, screening algorithm and clinical management. We conducted electronic search of potential eligible studies published in English in the Cochrane Controlled Register of Trials, PubMed, Medrxiv, Google scholar and Clinical Trials Registry databases. We included case studies, case series and observational and which MERS-CoV and COVID-19 to HIV/TB TB in adult We screened titles, abstracts and full articles As we heterogeneity in results reported

the pandemic and the number of new COVID-19 cases expected to rise in the next few months. The intersecting coronavirus, TB and HIV epidemics in sub-Saharan African countries where HIV and TB have the highest prevalence and incidence respectively, pose many challenges from the point of view of COVID-19/TB diagnostics, COVID-19/HIV/TB clinical management and post COVID-19 epidemic TB incidence.
In fact, the pathogenicity of COVID-19 could be accelerated in people living with HIV with compromised immunity [1]. Recent evidence has indicated a substantial association between coronavirus-related Lower Respiratory Tract Infections (LRTIs) and increased risk of death in immune-compromised individuals [15][16]. At the same time, the depletion of CD4 T cells in HIV and latent TB-infection disrupts the integrity and architecture of TB granulomas in the lung, thus facilitating progression to active TB [17,[18][19]. Similarly, TB promotes a microenvironment which facilitates the replication of HIV-1 via various mediators [20]. In fact, irreversible improvements in the lung architecture after SARS-CoV and/or TB play a signi cant role in both SARS-CoV and TB pathogenesis. Nonetheless, severe SARS-CoV can induce the development of rapid pulmonary brosis compared with mild courses of SARS-CoV disease usually advanced to organize phase diffuse alveolar damage (DAD) and eventual long-term deposition of brous tissue [21]. On the whole, SARS-CoV, HIV and TB co-infection may have deleterious consequences in all stages of SARS, HIV and TB because the triple pandemics are related in the immuno-pathological phase, constituting a vicious circle. A thorough understanding of the interactions between the three deadly pandemics is crucial. Reviewing the statistics in relation to high burden HIV/TB countries and recent World Health Organization data on COVID-19 in Sub-Saharan Africa; the following countries may expect an increased number of TB during or post COVID- 19 [7,[9][10][11][12][13][14][15][16][17][18][19][20][21][22]. The map was drawn to illustrate the distribution of COVID-19, HIV and TB in the nine high burden countries in Sub-Saharan African (Fig. 1). The aim of this study was to review different studies on SARS-CoV or MERS-CoV associated with HIV/TB co-infection or only TB and understanding the interactions between HIV, TB and COVID-19 and its implications on the burden of the COVID-19 among TB/HIV patients, screening algorithm and management.

Methods
The protocol was accepted by the international prospective register of systematic reviews (PROSPERO) (identi cation number: CRD42020181457). We conducted a systematic review of the literature to examine SARS-CoV or MERS-CoV associated with HIV/TB or TB co-infection. The justi cation to conduct a systematic review rather than a meta-analysis was the anticipated heterogeneity in the literature. We utilized formal methods of literature search, selection of articles for inclusion, abstraction of data and quality assessment, and synthesis of results to review the literature on to examine SARS-CoV or MERS-CoV associated with HIV/TB or TB co-infection. Furthermore, we computed the test of two proportions with STATA version 14 to compare SARS, MERS or COVID-19 disease severity compared to TB and/or HIV.

Inclusion Criteria
The inclusion criteria were studies published in English, from January 2020 until May 2020 that established co-occurrence of SARS-CoV, MERS-CoV, COVID-19 HIV and TB. Study designs included case reports, case series and observational studies (case-control, prospective and retrospective cohorts).
Studies reporting COVID-19/HIV co-infection without screening PTB, those reporting other outcomes, letters to the editor, theoretical and incomplete studies were excluded. The outcomes include TB occurrence (before, during or after SARS, MERS or COVID-19), SARS, MERS or COVID-19 severity (mild, moderate, severe and critical stages) in case of HIV/TB or TB co-infections, the mean time of COVID-19 severe/critical stages occurrence and the fatality rate.

Search Strategy
We searched eligible studies from 01 January 2002 to 21 May 2020 through Medline (PubMed), Google Scholar, Medrxiv and the Cochrane Library without any study design, published in English. Additionally, the WHO COVID-19 database [23] and Clinicaltrials.gov were also used to search for ongoing and completed studies related to co-infection COVID-19/HIV/TB. The following terms were used "SARS-CoV", "MERS-CoV", "COVID-19", "SARS-CoV-2" AND "pulmonary tuberculosis", "PTB", "lung TB", "TB" AND "HIV/TB co-infection" AND "TB/SARS co-infection" AND "TB/MERS co-infection "TB/Covid-19 co-infection" AND "HIV/SARS co-infection" AND "HIV/MERS co-infection AND "HIV/COVID-19 co-infection". Relevant articles published in English that resulted from the searches, and references cited therein, were reviewed and duplicate studies were removed. After removing duplicates, we checked the title and abstract, and reviewed full-text, inclusions and exclusions were recorded following PRISMA guidelines presented in the form of a PRISMA ow diagram and detailed reasons recorded for exclusion. Critical appraisal checklists appropriate to each study design were applied and checked by a second team member.

Results
Electronic search identi ed 315 articles. Inclusions and exclusions are reported following PRISMA guidelines presented in the form of a PRISMA ow diagram ( Fig. 2)    two patients were reported in the Kingdom of Saudi Arabia and one patient was reported in India. The rst case was 48 years (male), the second 18 years (male), the third 20 years (male) [24], the forth 54 years (male), the rth 39 years [25], the sixth 30 years (male) [26], the seventh 13 years (female), the eighth 30 years (female) [27], the ninth 76 years old (female) [28], the tenth 26 years (male), the eleventh 67 years (male) and the twelfth 76 years (male) [29]. Table 2 describes all cases. For further clari cations, cases were grouped as follows: Previous PTB diagnosed with SARS-CoV or MERS-CoV The second, third, seventh, eighth, tenth, the eleventh and twelfth cases were known to have a history of PTB (sputum smear-negative for acid-fast bacilli) and became infected with SARS-CoV ( rst, second cases, tenth, the eleventh and twelfth) or MERS-CoV(seventh and eighth cases). PTB diagnosis was made based on previous exposure to TB, relevant symptoms of typical PTB, chest radiographs suggestive of active disease. SARS-CoV or MERS-CoV was con rmed based on ampli cation of SARS-CoV/MERS-CoV RNA by reverse transcriptase-polymerase chain reaction (RT-PCR) from sputum. Both the second and third cases were managed with corticosteroids and anti TB drugs. Clinical management was not speci ed to the seventh and eight cases; however anti TB drugs were administered. Lopinavir/r, Arbidol, methyl prednisolone, empirical antibiotics, traditional Chinese medicine and antituberculosis treatment were indicated to the tenth, eleventh and twelfth cases. Five out of seven had severe/critical COVID-19 and had a long recovery process.
Newly PTB diagnosed with SARS-CoV The rst, sixth and ninth cases were diagnosed with PTB (positive acid-fast bacilli smear on sputum samples) while they were admitted for SARS-CoV in the hospital and RT-PCR was used to con rm SARS-

Observational studies
We included three observational studies (a case control and two retrospective studies). The case control and the rst retrospective studies were conducted in China [30][31] and the last retrospective cohort was undertaken in eight countries (Italy, Belgium, Brazil, France, Russia, Singapore, Spain, Switzerland) [32] (see Table 2). The proportion of severe/critical SARS, MERS and COVID-19 cases with HIV/TB or TB co-infection was higher than that in the mild/moderate stages (P = 0.0008). The onset of COVID-19 severe/critical stages the mean of 3.4 days [30] and the median of 9 days [32] for two observational studies and 10 days for a case study [29]. Only one observational study reported the fatality rate, meaning COVID-19/TB coinfection case fatality rate was 40% in this study.

Discussion
Reviewing the above case reports grouped in three, the interactions between SARS-CoV, HIV and TB have illustrated PTB may occur during SARS-CoV or after SARS. It is highly likely that both cases 4 and 5 acquired active pulmonary tuberculosis after contracting SARS, as both had laboratory-con rmed clinical syndromes associated with SARS, and both recovered well without anti-TB treatment, with initial biochemical and radiological improvement [24]. The analysis of cases found that SARS-CoV could induce a transient suppression of cellular immunity that further predisposed patients to exacerbated reactivation or new TB infection, as is the case with HIV. SARS-CoV and HIV may decrease conjunctly CD4 count and lymphocytes, adding high corticosteroids [24] as treatment for SARS-CoV in cases 4 and 5 may be TB precipitant factors. Following this, SARS-CoV patients may be more susceptible to active and latent TB during SARS-CoV infection as evidenced by the rst, sixth and ninth cases or after SARS-CoV infection as in cases 4 and 5. It is important to realize that lung lesions due to SARS and/or TB may increase signi cantly the likelihood of SARS-CoV and TB.
The overall review included 46 cases among whom 31(67.4%) had severe/critical SARS, MERS or COVID-19, 15(32.6%) had mild to moderate. We computed two proportions between severe/critical and mild/moderate with STATA 14. The percentage of severe/critical SARS, MERS and COVID associated with TB was higher than those with mild/moderate stages (P= 0.0008). Among severe/critical stages, 60.86% had TB past medical history. These ndings are a cautionary reminder to clinicians that TB infection status should be considered when treating COVID-19 patients in order to prevent rapid deterioration in patient health [30]. The onset of COVID-19 severe/critical stages varied between studies with the mean of 3.4 days [30] and the median of 9 days [32], this illustrated that the progression of COVID-19 disease may be faster and more severe [30]. The fatality rate of 40% among COVID-19/TB or HIV/TB co-infected patients should be considered with caution because only one study reported the fatality rate, with a poor study design and small sample size. However, COVID-19/TB or HIV/TB co-infections fatality seems to be higher than fatality rate of 5.96% worldwide [7]. Particular emphasis is placed on two cases with SARS-CoV/HIV and TB co-infection; given his immune-compromised condition, the rst patient SARS-CoV/HIV and TB co-infection went on a relatively mild course and the second developed COVID-19 critical stage and died. The rst SARS-CoV/HIV/TB case had mild SARS-CoV because of two possible reasons. Firstly, during the viraemic process, the antiretroviral therapy (ART) regime may have protective antiviral effects [26]. Kaletra was found to have some in-vitro anti-coronavirus activities [26]. Even though ART regimen was not given for the second case of SARS-CoV/HIV/TB, this case had multiple comorbidities including hepatitis B, metastatic prostate cancer and liver cirrhosis [32].

TB diagnostic in COVID-19/HIV co-infection
Suspected cases of COVID-19 and TB show similar fever and/or respiratory symptoms (di cult respiration, coughing, chest pain, etc.). COVID-19 RT-PCR should be done in real-time for differential diagnosis of cases with unknown respiratory syndromes such as PTB [33]. Due to poor outcomes among COVID-19/HIV/TB or COVID-19/TB co-infections, we recommend COVID-19 real-time RT-PCR should be coupled with Xpert MTB/RIF assay. In suspected HIV/TB co-infected patients, Xpert MTB/RIF should be used rst rather than traditional microscopy, culture and drug susceptibility testing (DST) [33]. Instead of collecting upper respiratory tract specimens, lower respiratory tract specimens, such as sputum, bronchoalveolar lavage, and tracheal aspirates should be collected in suspected COVID-19/HIV/TB or COVID-19/TB co-infected patients. COVID-19 real-time RT-PCR may last at least 24 hours. At the same time, the Xpert MTB / RIF assay detects M. tuberculosis and rifampicin resistance within less than two hours [34]. Xpert MTB/RIF is also a major advance in the diagnosis of TB, particularly for multidrugresistant (MDR) TB and HIV-associated TB [34]. The Xpert MTB/RIF assay simultaneously detects M. tuberculosis and rifampicin resistance in less than two hours [34]. Furthermore, Xpert MTB/RIF is a major advance for TB diagnostic; especially for MDR-TB and HIV/TB co-infection [34]. The Xpert MTB / RIF assay's sensitivity to detect TB is superior to that of microscopy and comparable to that of solid culture, along with high speci city [35]. positive, HIV viral load and CD4 count. All people with cough of any duration, fever, short breathing, sore throat, weight loss, hemoptysis, night sweat, arthralgia or myalgia should be investigated for TB. Xpert MTB/RIF assay coupled with COVID-19 IgG/IgM should be indicated. A recent study has found that the speci cities of serum IgM and IgG to diagnose COVID-19 were both more than 90% when compared to molecular detection [37]. If the Xpert MTB/RIF assay is negative, see options 2 and 3.

Drug-drug interactions and clinical considerations
In the case of concurrent HIV and tuberculosis infection plus SARS-CoV-2 infection, additional drug might cause interaction complicating the integrated therapy. In fact, some pharmaceutical interventions found for COVID-19 treatment including Protease inhibitors (PIs) ( atazanavir, lopinavir, ritonavir, duranavir, raltegravir,cobicistat), remdesivir, chloroquine, hydroxychloroquine, methylprednisolone, anticoagulants and carrimycin may interfere and interact to TB and/or HIV treatments in multiple ways. Although protease inhibitors (PIs) were developed to be selective inhibitors of HIV-1 replication, they have shown inhibitory activity against a wide variety of pathogens [38], including SARS-CoV. Lopinavir / ritonavir (LPV/r) has a moderate anti-SARS-CoV-2 antiviral activity which works against the 3CL protease virus [39][40]. A recent systematic review concluded that it is unclear whether LPV/r and other ART enhance clinical outcomes in severe symptomatic disease or prevent infection in patients at high-risk of COVID-19 based on the evidence available [41], as most of the studies included were case studies and also observational studies were low of power. Drug-drug interactions between PIs and rifampicin are known in HIV/TB co-infection. Studies have demonstrated that co-administration of PIs with rifampicin reduces PIs systemic concentration to less than 75% (cytochrome P 450 induction) [42][43]. This may compromise COVID-19 treatment. Remdesivir should also not associate to rifampicin in COVID-19/TB co-infection because of strong induction [44]. A recent review has reported that Chloroquine phosphate and Hydroxychloroquine showed favorable outcomes in the recovery of COVID-19 patients [45,46,47,[48][49]. Both chloroquine and hydroxychloroquine are metabolized by hepatic cytochrome P450 enzyme 2D6 (CYP2D6) [50]. The most frequently involved in drug interactions are CYP3A4 and CYP2D6 [50]. The reduction in the e cacy of chloroquine when administered in conjunction with rifampicin may be due to the inducing effect of rifampicin on multidrug resistance associated protein (MRP) and development of CYP450 [51]. Additionally, high-dose chloroquine is more toxic than lower dose [44]. This is why; studies should clarify chloroquine and hydroxychloroquine dose adjustment in COVID-19/TB co-infection. Based on the above, dose adjustments should be taken into consideration in case PIs, chloroquine, hydroxychloroquine and remdesivir are administered with rifampicin. Another option is to shift rifampicin to rifabutin or adapted TB regimens without rifampicin. In contrast, clofazimine used in MDR-TB is a strong inhibitor of PIs, known substrates [52]. Then, caution should be taken when administered with PIs. Another TB drug with in vitro effect used in COVID-19 is carrimycin. Its use in COVID-19 may mitigate active TB and biases the TB diagnostic.
A study showed an association between corticosteroid use and lower mortality in COVID-19 patients [49].
Using a glucocorticoid in the early stages of the prognosis for a brief period of time could minimize the in ammation, but longer-term use could result in the risk of HIV and/or TB activation and even lack of treatment with TB. Careful use of corticosteroids with low-to-moderate doses in short courses is advised [49]. Besides, brosis and extensive pulmonary pathology secondary to TB and COVID-19, as de ned in the introduction, can reduce drug penetration at the lung sites. It is a signi cant risk factor for bad TB outcomes in the event of potential infection or reactivation of TB [53]. This may also induce MDR-TB or extensively drug-resistant tuberculosis (XDR-TB) or recurrent pneumonia. Then, special considerations should be taken into account in the clinical management of COVID-19/TB lung brosis. Some RCTs are currently underway evaluating the safety and effectiveness of anti brotic therapies on COVID-19 lung brosis [54].
Besides, liver and kidneys toxicities related to severe and critical COVID-19 need a tailored therapeutic approaches in HIV/TB co-morbidities due to some hepatotoxicity and nephrotoxicity of some HIV/TB drugs such as streptomycin, isoniazid, rifampicin, pyrazinamide, tenofovir disoproxil, atazanavir/ritonavir, lopinavir/ritonavir as well as HIV induced nephropathy and hepatitis associated to HIV.  [44] or lopinavir 400 mg/ritonavir 100 mg PO twice [44]. All of them should be associated with Azithromycin [44]. Drugs interactions should be reviewed as described above. Initial evaluation includes a chest x-ray, complete blood count (CBC), liver transaminases, renal function, in ammatory markers such as C-reactive protein (CRP), D-dimer, and ferritin, while not part of standard care, may have prognostic value.
Severe COVID-19 associated to HIV/TB co-infection: Hospitalized in COVID-19/TB unit as high-risk patients. Drug therapy and ventilator support are milestones. Clinicians can refer to COVID-19 antiviral therapy and immune-based therapy [44]. Start COVID-19 antiviral drugs as described in mild to moderate COVID-19, add immune-based therapy, initiate or continue anti TB drugs according to national guidelines and ART should be discontinued based on drug interactions and clinical considerations as described above. Remdesivir is recommended in severe/critical COVID-19 however this cannot be administered with rifampicin [44]. Short period low-dose corticosteroid therapy is preferred over no corticosteroid therapy in HIV/TB co-infection and also the patients are in the intensive care unit [44]. Anticoagulant therapy mainly with low molecular weight heparin should be initiated early as this appears to be associated with better prognosis in severe COVID-19 patients [55]. Ventilator support, oxygen through a face mask and symptomatic therapy should be indicated. Initial evaluation includes chest x-ray/CT-scan and CBC should be indicated. Liver transaminases and renal function should be monitored regularly in consideration of COVID-19/HIV/TB drug-drug interactions and clinical considerations. Measurements of in ammatory markers, D-dimer, and ferritin are part of the management.
Critical COVID-19 associated to HIV/TB co-infection: Hospitalized in COVID-19/TB unit with ICU as highrisk patients. Infection control and testing, ventilator support, hemodynamic, and drug therapy are milestones [44]. Apply COVID-19, TB and HIV management as described in severe COVID-19. Short period low-dose corticosteroid therapy, anticoagulant therapy and norepinephrine as the rst-choice vasopressor are recommended [44]. Anticoagulant therapy mainly with low molecular weight heparin appears to be associated with better prognosis in severe/critical COVID-19 patients with markedly elevated D-dimer [55]. There is strong evidence against the use of hydroxyethyl starches for the acute reanimation of adults with COVID-19 in shock [56]. In adults with COVID-19 in shock, if the peripheral oxygen saturation (SpO2) is < 92%, the review suggested starting supplemental oxygen if SpO2 is < 90% [56]. Initial evaluation includes chest x-ray/CT-scan and CBC should be indicated. Knowing that COVID-19 and TB may induce the development of severe lung disease leading to pulmonary brosis in the future, further studies are needed with cohorts of HIV/COVID-19 co-infected individuals. More research is needed to explore the effect of lung brosis related to COVID-19 in high burden HIV/TB countries. This pressing priority will shed light on the utility of prophylaxis treatments in preventing post-COVID-19 related LRTIs in high burden HIV/TB countries. Study Flow Chart