Role of vitamin C in treatment of community-acquired pneumonia in adult patients requiring hospitalisation: a systematic review protocol

Abstract

Introduction

Community-acquired pneumonia (CAP) is a leading cause of hospitalisation and is associated with a high mortality. Vitamin C is a powerful antioxidant and has been used in treatment of infections; however, its role as an adjunctive treatment in CAP is unclear. This review aims to assess the efficacy and safety of vitamin C in adults who require hospitalisation for CAP.

Methods and analyses

Searches will be conducted from inception to November 2023 on Ovid MEDLINE Daily and MEDLINE, Embase CINAHL, the Cochrane Central Register of Controlled Trials, Scopus, Web of Science and ClinicalTrials.gov databases with the aid of a medical librarian. We will include data from randomised controlled trials reporting vitamin C supplementation in patients with CAP requiring hospitalisation. Two independent reviewers will select studies, extract data and will assess the risk of bias by use of the Risk of Bias tool. The overall certainty of evidence will be assessed by use of the Grading of Recommendations Assessment, Development and Evaluation framework. Random-effects meta-analyses will be conducted, and effect measures will be reported as relative risks with 95% CIs.

Ethics and dissemination

No previous ethical approval is required for this review. The findings of this review will be submitted to a scientific journal and presented at an international medical conference.

PROSPERO registration number

483860.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This review proposes to analyse the beneficial effects of vitamin C among hospitalised patients with community-acquired pneumonia.

• The selection, data extraction and inclusion of studies will involve two independent reviewers.

• Robust statistical procedures will be employed to determine the certainty of evidence.

• Exclusion of studies of patients with mild pneumonia and treated in the community settings may limit the number found in the search.

Introduction

Community-acquired pneumonia (CAP) is defined as an acute infection of the pulmonary parenchyma acquired outside hospital and is a leading cause of morbidity and mortality worldwide. In the USA, CAP accounts for 4.5 million outpatient and emergency visits annually with 1.5 million requiring hospitalisation.¹ Worldwide, CAP is the second most common cause of hospitalisation and is the most common infectious cause of death.² Pneumonia remains a common terminal cause of death especially in older patients with multiple comorbidities.³ In Australia, influenza and pneumonia were found to be associated with a 33.5% (42.6 per 100 000 people) increased risk of mortality in 2022 compared with previous years.⁴ This increase was driven by more deaths due to COVID-19, which was the underlying cause for over a quarter (27%) of these deaths.⁴

CAP can be caused by a variety of bacteria, viruses and fungi. Streptococcus pneumoniae and respiratory viruses are the most frequently detected pathogens in patients with CAP.⁵ Inpatient mortality from CAP ranges between 4.2% and 5.5%.⁶ A recent cohort study which involved >7000 patients hospitalised for CAP in the USA found that 30-day mortality was 13% while mortality at 6 months was 23%.⁷ Recent studies^{8 9} suggest that despite advancements in clinical care, mortality rates from pneumonia have not any shown any substantial change over time. An excess inflammatory response seems to be partly responsible for treatment failure in some patients with CAP and has been associated with poor clinical response to antibiotics.¹⁰ Therefore, there is a need to explore adjunctive therapies that have immunomodulatory and barrier-enhancing function for treatment of CAP.

Vitamin C levels drop during acute inflammation/infection and patients with CAP have been reported to have low serum and leucocyte levels of vitamin C.¹¹ Vitamin C is a water-soluble vitamin with powerful antioxidant properties that can scavenge free radicals; however, it cannot be produced in humans because of lack of an enzyme called gluconolactone synthase.¹² This vitamin has immune-mediating properties as it has been found to support neutrophil migration to the site of infection and is responsible for production of hormones such as norepinephrine and vasopressin.^{12 13} These properties have led to an investigation of its potential role as an additional therapeutic agent in the treatment of pneumonia.

Previous randomised controlled trials (RCTs) evaluating the benefits of vitamin C supplementation among patients with pneumonia have yielded varied results. A systematic review¹⁴ in 2021 included five RCTs of oral and parenteral vitamin C supplementation for prevention and treatment of pneumonia. This review found a reduction in the number of days before improvement in oxygen saturation (1.03±0.16 days vs 1.14 days±1.0 days) and respiratory rate (3.61±1.50 days vs 4.04±1.62 days) in the vitamin C-treated group compared with the control group, but the certainty of evidence was limited by the small number of trials. A recent RCT,¹¹ which included 75 patients, evaluated intravenous vitamin C followed by oral treatment compared with placebo among patients with moderate to severe CAP and found no significant difference in 28-day mortality (0% vs 2%, p=0.49) but a trend towards reduced length of hospital stay (LOS) (median LOS 69 vs 121 hours, p=0.07) and time to clinical stability (22 vs 49 hours, p=0.08). Another RCT¹⁵ conducted on 872 critically ill patients with septic shock, one-third of whom had a respiratory source, yielded contrasting results. This study revealed that intravenous administration of vitamin C was linked to an elevated risk of death or persistent organ dysfunction at 28 days when compared with placebo (risk ratio 1.21, 95% CI 1.04 to 1.40, p=0.01). Consequently, the role of vitamin C in the treatment of CAP remains contentious and necessitates further elucidation.

Our systematic review will aim to assess the efficacy and safety of parenteral and/or oral vitamin C in combination with other therapies in adults with CAP.

Research question

What is the value of supplementing with vitamin C compared with placebo among patients who require hospitalisation for moderate to severe CAP?

Methods and analysis

This protocol adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols standards¹⁶ and has been registred in the International Prospective Registry of Systematic Reviews (no 483860) in November 2023.

Search strategy

With the help of a medical librarian, we will search the following electronic databases from inception to 17 November 2023: MEDLINE (from 1946), Embase (from 1974), CINAHL (from 1984), the Cochrane Central Register of Controlled Trials, Scopus, Web of Science and ClinicalTrials.gov. We will use a combination of keywords and medical subject headings as follows: adults, community acquired pneumonia, bronchopneumonia, lower respiratory tract infections, hospitalisation, Inpatients, Critical care, vitamin C, ascorbic acid, ascorbate, mortality, randomised controlled trials, and placebo intravenous administration, and oral vitamin C. No language restrictions will be applied.

We will not deliberately search grey literature, but plan to include any conference abstracts or unpublished data that were retrieved from our searches. This will include any unpublished data retrieved from the ClinicalTrials.gov. To maintain the current status of our systematic review, we will adopt a living status framework and will update our search every 3 months. The final search strategies for the academic databases are presented in online supplemental appendix 1.

Eligibility criteria

Design and population

We will include parallel-arm RCTs of adults aged ≥18 years with CAP. Pneumonia will be defined as symptoms of fever, dyspnoea, cough and sputum production along with imaging evidence of a pulmonary infiltrate requiring hospitalisation, including intensive care unit (ICU) admission. We will still include publications where authors have not clearly defined pneumonia but have used terms such as ‘pneumonia’ or ‘consolidation on imaging studies’ to define their target population. We will include studies which have enrolled patients with moderate to severe infection CURB-65 score¹⁷ ≥2 or Pneumonia Severity Index (PSI)¹⁸ ≥3, or those patients who were diagnosed with pneumonia and meeting the WHO definition of sepsis¹⁹ (systolic blood pressure <90 mm Hg; pulse rate >100/min, a respiratory rate >24/min and abnormal temperature (<36°C or >38°C)). In addition, we will also consider the following studies as eligible for inclusion if at least 80% of their patients meet our inclusion criteria for our population of interest and also consider those studies which include patients with severe organ failure including acute respiratory distress syndrome (ARDS)²⁰ as long as outcomes are reported for the population with severe pneumonia as a risk factor for ARDS.

Intervention

Clinical trials with at least one arm involving the administration of parenteral and/or oral vitamin C alone or in combination with other micronutrients and therapies will be included. We will include studies that administered very small doses of vitamin C, which are typically 100 mg/day and are used in parenteral nutrition as well as studies using mega doses >12 g/day. Studies will be divided into three dose ranges (low dose range <6 g/day, moderate dose range (6–12 g/day) and high dose range (>12 g/day). These cut-off points are arbitrary and previous research²¹ on adverse effects of vitamin C have typically used 6 g/day as the lower limit of high doses for vitamin C. If studies have used a weight-based regimen, then we will use the mean patient body weight to determine total daily dose of vitamin C. If mean body weight is not provided, then we will assume 70 kg as the mean weight to determine daily vitamin C doses. We will not exclude studies on the basis of administered cointerventions including other vitamins and antibiotics.

Comparator arm

We will only include studies which have at least one control arm which included patients who were not prescribed parenteral and/or oral vitamin C. The control arm may receive placebo or any other active treatment. We will exclude studies that compare different regimens of parenteral and/or oral vitamin C without an arm that lacks vitamin C administration.

Types of outcome measures

Primary outcomes

The primary outcome will include 30-day and 90-day mortality from date of admission including in-hospital deaths. We will include mortality data closest to the time points of interest.

Secondary outcomes

The secondary outcomes will include:

1. The need for an admission to the ICU and duration of ICU stay.

2. Use of non-invasive and invasive ventilation.

3. Use of vasopressor support.

4. The time to clinical stabilisation measured on a WHO ordinal scale²² for change in clinical status or other severity measures such as the reduction in number of days before improvement in hypoxaemia or respiratory rate.

5. The duration of hospitalisation.

6. 30-day readmission rates.

7. Development of stage 3 acute kidney injury as determined by the Kidney Disease: Improving Global Outcomes Criteria²³ or/and the need for dialysis.

Other secondary outcomes will include serious adverse events resulting in discontinuation of vitamin C treatment, new haemolysis, hypoglycaemic events, nephrolithiasis or development of clinically significant oxalosis. The time points for all secondary outcomes will be within 30 days of the follow-up period from time of admission or the closest available time points of interest.

Study selection

Two reviewers will screen identified citations at the title and abstract screening level using predefined eligibility criteria electronically by use of reference manager. Potentially eligible citations will then be reviewed at the level of full-text screening by paired reviewers. The screening will be completely independent and in duplicate, and any disagreements will be resolved by involvement of a third reviewer. Studies will only be included if meeting eligibility criteria and reporting at least one primary or secondary outcome of interest.

Data extraction and risk of bias assessment

Data will be extracted on study populations, interventions, comparators, and outcomes of interest and we will conduct prespecified subgroup analyses. We will specifically extract data on the following variables:

1. Study-level characteristics including publication status, study design and funding type (institutional or pharmaceutical).

2. Patient-level baseline demographics: country, mean or median age, gender, proportion of patients with chronic kidney disease, congestive heart failure, cardiovascular disease, chronic lung disease (bronchial asthma, chronic obstructive pulmonary disease, interstitial lung disease and bronchiectasis), diabetes mellitus, hypertension, smoking status, use of alcohol or illicit drugs.

3. Patient-level baseline clinical characteristics: type of pneumonia (community or hospital acquired), illness severity score as determined by CURB-65 or PSI, proportion of patients requiring ICU support, proportion of patients needing dialysis, proportion of patients requiring vasopressors, proportion of patients needing basic oxygen modalities (nasal prongs, face masks, high-flow oxygen, non-invasive ventilation and invasive mechanical ventilation), mean or median lactate levels.

4. Study-level intervention characteristics: description of regimen used, number of patients randomised, allocation concealment, total daily vitamin C dose, duration of treatment, use of cointerventions (thiamine, corticosteroids, type of antibiotics used) with associated doses and duration of treatment, mean time from presentation or enrolment to administration of the intervention, mean volume of fluid administered from admission to randomisation and mean volume of fluid administered up to 24 hours post-randomisation.

5. Study-level binary outcome variables: number of outcomes reported, follow-up time in days and number of participants analysed.

6. Study-level continuous outcomes: outcomes reported, follow-up time in days, number of participants analysed, measures of and estimates of central tendency and variability as mean (SD) or median (IQR).

7. Subgroup analyses as reported in sensitivity analyses.

8. Outcome-level risk of bias (RoB) assessment for dichotomous and continuous outcomes with assessment of domain and justification.

RoB assessment

Two reviewers will extract data and evaluate the RoB using the modified version of RoB tool,²⁴ which has been extensively employed in previous research. RoB will be classified as ‘low’, ‘probably low’, ‘probably high’ or ‘high’ for the following domains: bias because of randomisation, bias due to deviations from the intended intervention, bias due to missing outcome data, bias in the selection of the reported results, bias in measurement of the outcomes and other biases. The overall RoB will be rated based on the highest risk attributed to any of the above-mentioned criteria.

Assessment of the certainty of the evidence

We will use the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework²⁵ to determine the overall certainty of the evidence. This evaluation will be based on the following domains: RoB assessment, imprecision, indirectness, inconsistency and publication bias. The overall certainty of evidence will be rated as ‘very low’, ‘low’, ‘moderate’ or ‘high’. The certainty of evidence for the RoB domain will be rated down for studies displaying lack of blinding for subjective outcomes only. In addition, we will make judgements on imprecision by using the minimally contextualised approach as suggested by Hultcrantz et al.²⁶ We will also consider any CI confining the null effect to be imprecise while taking into consideration important and trivial effects.²⁶

Statistical analyses

We will use random-effects modelling for our meta-analyses. All statistical analyses will be performed by use of Stata software V.18.0. We will present binary outcomes as relative risks and continuous outcomes as mean differences or standarised mean differences with 95% CIs. We will assume a normal distribution for continuous outcome variables and will convert medians and IQRs to means and SDs as suggested by the Cochrane Collaboration Group.²⁷ The statistical heterogeneity among studies will be assessed by use of the I² measure,²⁸ and inconsistency will be judged by GRADE assessments on the basis of the magnitude and direction of heterogeneity. Publication bias will be assessed by visual inspection of funnel plots. Furthermore, this method will be supplemented by use of a regression test if at least 10 studies are available.

Sensitivity analyses

If sufficient data are available, we will plan subgroup analyses: to determine whether higher doses of vitamin C have incremental effects (high-dose vitamin C vs moderate-dose vitamin C vs low-dose vitamin C; hypothesis: greater benefits with incremental doses), vitamin C as monotherapy versus vitamin C administration with other micronutrients (hypothesis: no difference in effects), treatment benefits of patients with or without septic shock (hypothesis: greater benefits among patients with septic shock), any differential benefits of vitamin C supplementation depending on patients’ influenza or COVID-19 status, and based on demographic factors such as age, gender and smoking status. We plan to limit sensitivity analyses to peer-reviewed studies available in full text and with a low RoB.

Patient and public involvement

We do not plan to involve patients or consumers directly in the conduct of this systematic review.

Ethics and dissemination

Patients with CAP who require hospitalisation have significant morbidity and mortality worldwide.² Given publication of recent clinical trials,¹¹ an updated evidence synthesis regarding the efficacy and safety of different vitamin C regimens for CAP is required. The findings of this systematic review will inform hospital clinical practices for the management of patients admitted with CAP.

This systematic review will have a number of strengths. First, we intend to summarise evidence relevant to patients who have a greater severity of pneumonia that warrants hospital admission. This is in contrast to recent reviews^{14 29} on vitamin C which have focused on prevention and treatment of pneumonia in general. Second, we attempt to include recent clinical trials that have been published recently resulting in an updated review. This will lead to higher certainty estimates for our primary and secondary clinical outcomes of interest. Third, we intend to use a comprehensive search strategy with the aid of a medical librarian and will incorporate both published and unpublished literature. This systematic review will use robust statistical analysis and will perform predefined subgroup analyses and intends to use the GRADE methodology to systematically evaluate certainty in effect estimates. We also intend to share our data with other living network meta-analyses teams so that the evidence on vitamin C supplementation can be used iteratively for indirect comparisons with other treatments for patients with CAP.

This protocol describes the detailed methodology for a planned systematic review and meta-analysis that aims to address comparative efficacy and safety of vitamin C supplementation as an adjunct treatment for hospitalised patients with CAP. The results of this review will be of importance to respiratory and generalist physicians who are involved in care of patients with CAP. No previous ethical approval will be required for this review. We plan to disseminate the study results through national and international conferences and publish our findings in a peer-reviewed journal.

Ethics statements

Patient consent for publication

Not applicable.