<i>Aloe vera</i> for prevention and treatment of infusion phlebitis

Guo Hua Zheng; Liu Yang; Hai Ying Chen; Jian Feng Chu; Lijuan Mei

doi:10.1002/14651858.CD009162.pub2

Aloe vera para la prevención y el tratamiento de la flebitis por infusión

Authors' declarations of interest

Version published: 04 June 2014 Version history

https://doi.org/10.1002/14651858.CD009162.pub2

Collapse all Expand all

Resumen

available in

Antecedentes

Hasta el 80% de los pacientes hospitalizados recibe tratamiento por vía intravenosa en algún momento durante el ingreso. Alrededor de entre el 20% al 70% de los pacientes que reciben tratamiento por vía intravenosa desarrolla flebitis. La flebitis por infusión se ha convertido en una de las complicaciones más frecuentes en los pacientes con tratamiento por vía intravenosa. Sin embargo, los efectos de los tratamientos habituales como la aplicación externa de alcohol al 75% o sulfato de magnesio del 50% al 75% (MgSO₄) no son satisfactorios. Por lo tanto, existe la necesidad urgente de desarrollar nuevos métodos para prevenir y aliviar la flebitis por infusión.

Objetivos

Evaluar sistemáticamente los efectos de la aplicación externa de Aloe vera para la prevención y el tratamiento de la flebitis por infusión asociada a un dispositivo de acceso intravenoso.

Métodos de búsqueda

El coordinador de búsquedas de ensayos del Grupo Cochrane de Enfermedades Vasculares Periféricas (Cochrane Peripheral Vascular Diseases Group) buscó en el registro especializado (última búsqueda febrero de 2014) y en CENTRAL (2014, número 1). Además, el coordinador de búsquedas de ensayos efectuó búsquedas en MEDLINE hasta la semana 5 de enero de 2014, EMBASE hasta la semana 6 de 2014 y AMED hasta febrero de 2014. Los autores de la revisión buscaron en las siguientes bases de datos chinas hasta el 28 de febrero de 2014: Chinese BioMedical Database (base de datos biomédica chica), Traditional Chinese Medical Literature Analysis and Retrieval System (Sistema de recuperación y análisis de literatura médica china tradicional), China National Knowledge Infrastructure (Infraestructura de conocimiento nacional china), Chinese VIP Information (Información VIP china), Chinese Academic Conference Papers Database y Chinese Dissertation Database (bases de datos de artículos de conferencias académicas y disertaciones chinas) y China Medical Academic Conference (Conferencia académica médica china). Se buscó en las bibliografías de las publicaciones localizadas y relevantes. No hubo restricciones con respecto a la fecha o al idioma de publicación.

Criterios de selección

Se incluyeron ensayos controlados aleatorizados (ECA) y ensayos controlados cuasialeatorizados que reclutaran participantes que recibieron Aloe vera tópico o productos derivados de Aloe vera en el sitio de punción de la piel, con o sin tratamiento habitual en el mismo sitio.

Obtención y análisis de los datos

Dos autores de la revisión de forma independiente extrajeron los datos sobre las características de los estudios, la descripción de la metodología y los desenlaces de los ensayos elegibles y evaluaron la calidad de los estudios. Los datos fueron analizados mediante RevMan 5.1. Para los desenlaces dicotómicos, los efectos se calcularon mediante razón de riesgos (RR) con el intervalo de confianza (IC) del 95%. Para los resultados continuos, se utilizaron las diferencias de medias (DM) con los IC del 95% para calcular los efectos.

Resultados principales

Se identificaron 43 ensayos (35 ECA y ocho ensayos controlados cuasialeatorizados) con 7465 participantes. Veintidós ensayos con 5546 participantes analizaron la prevención de la flebitis con Aloe vera, y 21 ensayos adicionales con 1919 participantes estudiaron el tratamiento de la flebitis. Los estudios incluidos compararon la aplicación externa de Aloe vera solo o con intervenciones sin Aloe vera con ningún tratamiento o las mismas intervenciones sin Aloe vera. La duración de la intervención duró de uno a 15 días. La mayoría de los estudios incluidos fue de calidad metodológica baja con dudas acerca del sesgo de selección, el sesgo de desgaste, el sesgo de informe y el sesgo de publicación.

Los efectos de la aplicación externa de Aloe vera fresca sobre la prevención de la incidencia total de flebitis variaron entre los estudios y no se combinaron los datos. Aloe vera redujo la aparición de flebitis de tercer grado (RR 0,06; IC del 95%: 0,03 a 0,11; p < 0,00001) y flebitis de segundo grado (RR 0,18; IC del 95%: 0,10 a 0,31; p < 0,00001) en comparación con ningún tratamiento. En comparación con la aplicación externa de alcohol al 75%, o de MgSO₄ al 33%, el Aloe vera redujo la incidencia total de flebitis (RR 0,02, IC del 95%: 0,00 a 0,28, p = 0,004 y RR 0,43, IC del 95%: 0,24 a 0,78, p = 0,005, respectivamente), pero no había evidencia clara de que se produjera un efecto cuando se comparó con MgSO₄ al 50% o 75% (incidencia total de flebitis RR 0,41, IC del 95%: 0,16 a 1,07, p = 0,07 y RR 1,10, IC del 95%: 0,54 a 2,25, p = 0,79, respectivamente; flebitis de tercer grado (RR 0,28, IC del 95%: 0,07 a 1,02, p = 0,051 y RR 1,19, IC del 95%: 0,08 a 18,73, p = 0,9, respectivamente; flebitis de segundo grado RR 0.68, IC del 95%: 0,21 a 2,23, p = 0,53 en comparación con el 75% de MgSO₄), salvo una reducción de la flebitis de segundo grado cuando se comparó elAloe vera con el 50% de MgSO₄ (RR 0,26, IC del 95%: 0,14 a 0,50, p < 0,0001).

En el tratamiento de la flebitis, el Aloe vera fue más eficaz que MgSO₄ al 33% o 50% en cuanto a cualquier mejoría (RR 1,16; IC del 95%: 1,09 a 1,24; p < 0,0001 y RR 1,22; IC del 95%: 1,16 a 1,28; p < 0,0001, respectivamente) y mejoría marcada de la flebitis (RR 1,97; IC del 95%: 1,44 a 2,70; p < 0,001 y RR 1,56; IC del 95%: 1,29 a 1,87; p = 0,0002, respectivamente). En comparación con MgSO₄ al 50%, el Aloe vera también mejoró las tasas de recuperación de la flebitis (RR 1,42; IC del 95%: 1,24 a 1,61; p < 0,0001). En comparación con tratamientos habituales como la aplicación externa de hirudoide, mucopolisacárido de ácido sulfónico y dexametasona utilizados solos, añadir Aloe vera mejoró la recuperación de la flebitis (RR 1,75; IC del 95%: 1,24 a 2,46; p = 0,001) y tuvo un efecto positivo sobre la mejoría general (mejoría marcada RR 1,26; IC del 95%: 1,09 a 1,47; p = 0,0003; cualquier mejoría RR 1,23; IC del 95%: 1,13 a 1,35; p = < 0,0001). El Aloe vera sola o en combinación con el tratamiento habitual fue más efectiva que el tratamiento habitual solo para mejorar los síntomas de la flebitis, incluido el acortamiento del tiempo de eliminación de los síntomas de tumefacción y enrojecimiento, el tiempo de alivio del dolor en la localización de la vena de infusión y el tiempo de resolución de la flebitis. Otros desenlaces secundarios que incluyen la calidad de vida relacionada con la salud y los efectos adversos no se informaron en los estudios incluidos.

Conclusiones de los autores

No existe evidencia sólida para la prevención o el tratamiento de la flebitis por infusión con la aplicación externa de Aloe vera. La evidencia disponible actual está limitada por la calidad metodológica deficiente y el riesgo de informe selectivo de los desenlaces de los estudios incluidos, y por la variación en el tamaño del efecto entre los estudios. Por lo tanto, los efectos positivos observados con la aplicación externa de Aloe vera para prevenir o tratar la flebitis por infusión en comparación con ninguna intervención o la aplicación externa de MgSO₄ al 33% o 50% se deben observar con precaución.

PICOs

Population

Intervention

Comparison

Outcome

The PICO model is widely used and taught in evidence-based health care as a strategy for formulating questions and search strategies and for characterizing clinical studies or meta-analyses. PICO stands for four different potential components of a clinical question: Patient, Population or Problem; Intervention; Comparison; Outcome.

See more on using PICO in the Cochrane Handbook.

Resumen en términos sencillos

available in

Aloe vera para la prevención y el tratamiento de la flebitis por infusión

La flebitis por infusión es la inflamación aguda de una vena a causa de un tratamiento por vía intravenosa. En la práctica médica moderna, más del 80% de los pacientes hospitalizados recibirá tratamiento por vía intravenosa durante el ingreso y cerca del 20% al 70% de ellos puede desarrollar flebitis por infusión. Por lo tanto, la flebitis por infusión es la complicación más frecuente del tratamiento intravenoso. Sin embargo, los tratamientos habituales para la prevención o el tratamiento de la flebitis como la aplicación externa de alcohol al 75% o de sulfato de magnesio (MgSO₄) del 50% al 75% no son satisfactorios.

Esta revisión examinó 35 ensayos controlados aleatorizados y ocho ensayos controlados cuasialeatorizados con 7465 participantes. Veintidós ensayos con 5546 participantes analizaron la prevención de la flebitis con Aloe vera y 21 ensayos adicionales con 1919 participantes analizaron Aloe vera para el tratamiento de la flebitis. Los ensayos incluidos compararon principalmente la aplicación externa de Aloe vera fresca sola o con otro tratamiento sin Aloe vera como una compresa húmeda de alcohol al 75% o MgSO₄ al 33%, al 50% o al 75% con ningún tratamiento o con el mismo tratamiento sin Aloe vera. La duración de la intervención duró desde un día hasta 15 días. La mayoría de los estudios incluidos fue de calidad metodológica baja con dudas acerca del sesgo de selección, el sesgo de desgaste, el sesgo de informe y el sesgo de publicación. Se investigó la incidencia de flebitis con grados variables de gravedad, así como la tasa de resolución y el nivel de mejoría de la flebitis.

Las pruebas disponibles indican que la aplicación externa de Aloe vera fresca sola o combinada con otro tratamiento sin Aloe vera puede ser efectiva para la prevención y el tratamiento de la flebitis por infusión como resultado del tratamiento por vía intravenosa. Las conclusiones deben interpretarse con cautela debido a la baja calidad metodológica de los ensayos incluidos.

Authors' conclusions

Implications for practice

There is no strong evidence for the prevention or treatment of infusion phlebitis with external application of Aloe vera. The current available evidence is limited by the poor methodological quality and risk of selective outcome reporting of the included studies, and by variations in the direction and size of effect across the studies. The positive effects observed with external application of Aloe vera in preventing or treating infusion phlebitis compared with no intervention or external application of 33% or 50% MgSO₄ should therefore be viewed with caution.

Implications for research

Further research should consider the following points.

1. The methods of random sequence generation and allocation concealment should be adequately described.

2. Blinding of outcome assessors should be applied to reduce detection bias.

3. Clinical trial registers should be encouraged to provide the available protocols.

4. Participants who were withdrawals or dropouts during the trial should be clearly described.

5. Adverse events should be measured as a priority outcome.

Background

Description of the condition

Intravenous therapy is the giving of substances directly into a vein. Intravenous therapy may be used to correct electrolyte imbalances, deliver medications, for blood transfusion, or as fluid replacement. In modern medical practice up to 80% of hospitalised patients receive intravenous therapy at some point during their admission (Tager 1983; Tjon 2000). Compared with other routes of administration, the intravenous route is the fastest way to deliver fluids and medications throughout the body. The devices for intravenous therapy usually include a hypodermic needle, peripheral venous cannula or intravenous catheter and this is the most common invasive procedure among patients admitted to hospital (Waitt 2004). It is estimated that 200 million peripheral intravenous devices are placed each year in America alone (Maki 2008; Mermel 2001), while one in three inpatients have at least one peripheral venous catheter in situ (Boyd 2011). Medication, fluids, nutrition, and blood products can all be given via the peripheral intravenous route, and the plastic catheter may stay in situ long term. Although common, these practices are not devoid of complications, which may increase duration of hospital stay and admission cost (Waitt 2004). The complications that occur frequently in patients with peripheral intravenous therapy include infusion phlebitis, infection, infiltration or extravasation, fluid overload, hypothermia, electrolyte imbalance and embolism. Infusion phlebitis is in almost all cases a biochemical reaction to the mechanical irritation caused by the presence of the intravenous catheter (Maki 1991). The remaining cases may be chemical or infectious. Studies have shown that 20% to 70% of patients receiving peripheral intravenous therapy develop phlebitis (Hershey 1984; Monreal 1999; Tully 1981). In China, according to statistics, about 80% of the patients with intravenous therapy develop varying degrees of infusion phlebitis (Tang 2001).

Description of the intervention

Aloe or Aloe vera is a genus containing many species of flowering succulent plants with a rosette of large, thick, fleshy leaves (Klein 1988). Aloe vera species are frequently cultivated as ornamental plants both in gardens and in pots. Aloe has been known and used for centuries for its health, beauty and skin care properties. Two thousand years ago, the Greek scientists regarded Aloe vera as the universal panacea and the Egyptians called Aloe the plant of immortality (Surjushe 2008). It has a long history of use as an anti‐inflammatory herbal application for burns and for a variety of conditions in traditional medicine (Hutter 1996). Today, Aloe vera is still used as a herbal medicine. Aloe vera is used either internally or externally in humans and has some medicinal effects which have been supported by scientific and medical research (Vogler 1999). As a herbal medicine, Aloe vera juice is commonly used internally to relieve digestive discomfort (Langmead 2004) and externally to relieve skin discomforts such as minor burns, wounds and various skin conditions such as eczema and scabies (Feily 2009; Oyelami 2009; Visuthikosol 1995). The topical or external applications usually use aloe‐derived products, which include aloe vera cream, aloe vera mucilage, aloe vera gel etc. (Vogler 1999). However, crude aloe vera components such as fresh aloe vera leaf, fresh aloe vera stem, and aloe vera juice are also used as adjuvant treatment to treat skin disorders (Peng 2009; Zhang 2010).

How the intervention might work

Aloe vera contains 75 potentially active constituents including vitamins, enzymes, minerals, sugars, lignin, saponins, salicylic acids and amino acids (Shelton 1991), some of which have several pharmacological actions. These include the carboxypeptidase that inactives bradykinin, salicylates and substances that inhibit local vasoconstriction (Klein 1988). The anti‐inflammatory compound called C‐glucosyl chromone has been isolated from gel extracts (Hutter 1996). It has been proven that fresh Aloe vera can promote the attachment and growth of normal human cells in vitro and enhance the healing of wounded monolayers of cells (Winters 1981) whereby Aloe vera gel not only increased the collagen content of the wound but also changed the collagen composition and increased the degree of collagen cross linking (Chithra 1998). Some studies have shown that pure Aloe vera is effective in preserving skin circulation following frostbite injury (McCauley 1990) and in accelerating wound healing (Fulton 1990) and increasing the breaking strength of resulting scar tissue in patients (Heggers 1996). In addition, Aloe vera can stimulate fibroblasts, which produce the collagen and elastin fibres, making the skin more elastic and less wrinkled; and Aloe vera has cohesive effects on the superficial flaking epidermal cells by sticking them together, to soften the skin (West 2003). To summarise, the mechanism of action of Aloe vera includes healing properties, anti‐inflammatory activity, effects on the immune system, moisturising and anti‐aging effects and antiseptic effects (Surjushe 2008). Aloe may therefore be beneficial for the prevention and treatment of infusion phlebitis.

Why it is important to do this review

Infusion phlebitis is acute inflammation of a vein in the presence of and directly connected to an intravenous access device. It may be influenced by the pH and osmotic pressure of the solution, the size of the vein used, the size and material of the catheter, and the infusion period (Kuwahara 1998); and may be the most common complication of intravenous therapy. Especially during cancer chemotherapy, the toxicity of and irritation caused by the chemotherapy drugs can directly lead to phlebitis and even tissue necrosis. Yoh et al reported that the incidence of phlebitis after a six‐minute infusion of vinorelbine was 16% to 33% (Yoh 2007). Phlebitis not only brings great physiological and psychological suffering to the patients but also reduces the treatment effect and prolongs the process of treatment. The infusion phlebitis may initially manifest itself in the form of pain, erythema and swelling, warmth, hardening and thickening of the skin in the injection area (Schmid 2000). However, it can quickly develop into more serious symptoms such as severe local tissue ulceration and necrosis, even leading to serious limb dysfunction, and could ultimately cause thrombus formation and pyrexia. This is most likely to be caused by the pH and osmotic pressure of the solution, the mechanical irritation of the catheter or chemical irritation of the drugs (Emiko 2009). Therefore, infusion phlebitis does not only increase the psychological burden of the patient it also affects the quality of care. The routine or traditional methods for prevention and treatment of infusion phlebitis are varied, for example self care, anti‐inflammatory medication, cortisone preparations, rubbing or flushing the site with 75% alcohol or 0.9% saline solution, and compresses with 50% to 75% magnesium sulphate, but none are completely effective (Curran 1990; Nakayama 2002; O'Grady 2002). There is an urgent need to develop new methods to prevent and alleviate infusion phlebitis.

In clinical settings in China, it is common for Aloe vera to be externally applied to prevent and treat infusion phlebitis (Cao 2008; Hu 2006; Lu 2008; Yang 2008). Studies on its effectiveness have small sample sizes and the real effect is difficult to confirm. This systematic review of randomised and quasi‐randomised trials was conducted to clarify its efficacy.

Objectives

To systematically assess the effects of external application of Aloe vera for the prevention and treatment of infusion phlebitis associated with the presence of an intravenous access device.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) and quasi‐randomised controlled trials (qRCTs) which evaluated Aloe vera used alone or in combination with other treatments for infusion phlebitis were included irrespective of blinding, publication status and language. We excluded non‐randomised studies. Randomised cross‐over trials and cluster‐RCTs were eligible.

Types of participants

We included participants, regardless of gender, age or ethnic origin, who suffered from phlebitis of a peripheral limb vein that was associated with the presence of an intravenous access device, or who were at risk of developing phlebitis because of insertion of an intravenous access device. Peripheral limb veins are frequently used for siting intravenous access devices, including the dorsal venous network of the hand, veins of the forearm, veins of the cubital fossa, and dorsalis pedis veins.

Types of interventions

We included trials that compared topical external application of any type of Aloe vera (fresh Aloe vera slice or fresh Aloe vera juice) or aloe‐derived products, or Aloe vera plus non‐Aloe vera treatments, at the site of punctured skin with no treatment, routine treatment or the same non‐Aloe vera treatment at the same site. We excluded trials without a definite duration of treatment. In addition, we excluded those trials in which Aloe vera plus other non‐Aloe vera treatments were used in the treatment group and the same non‐Aloe vera treatments were not used in the control group.

Types of outcome measures

Primary outcomes

1. The incidence of phlebitis (for preventive effect) as assessed by the total incidence of phlebitis. We also assessed the incidence of third degree and second degree phlebitis.

2. The rate of resolution of phlebitis (for treatment effect) as assessed by the total improvement rate of phlebitis. We also assessed the recovery rate and marked improvement rate of phlebitis.

Secondary outcomes

1. The duration of successful placement of a venous catheter, defined as the length of time which the venous catheter, such as the needle, was retained in the vein after infusion.

2. Symptom improvement of phlebitis.

3. Health‐related quality of life, using quality of life measures as reported by the study authors.

4. Adverse effects: any adverse events as a result of the use of Aloe vera for treatment or prevention, such as toxic response, anaphylaxis, etc.

Search methods for identification of studies

There were no restrictions on the basis of date or language of publication.

Electronic searches

The Cochrane Peripheral Vascular Diseases Group Trials Search Co‐ordinator (TSC) searched the Specialised Register (last searched February 2014) and the Cochrane Central Register of Controlled Trials (CENTRAL) (2014, Issue 1), part of The Cochrane Library (www.thecochranelibrary.com). See Appendix 1 for details of the search strategy used to search CENTRAL. The Specialised Register is maintained by the TSC and is constructed from weekly electronic searches of MEDLINE, EMBASE, CINAHL, AMED, and through handsearching relevant journals. The full list of the databases, journals and conference proceedings which have been searched, as well as the search strategies used, are described in the Specialised Register section of the Cochrane Peripheral Vascular Diseases Group module in The Cochrane Library (www.thecochranelibrary.com).

In addition, the TSC searched MEDLINE (Appendix 2), EMBASE (Appendix 3) and Allied and Complementary Medicine (AMED) (Appendix 4).

The following trial databases were searched by the TSC for details of ongoing and unpublished studies using the term "aloe vera":

World Health Organization International Clinical Trials Registry Platform (http://apps.who.int/trialsearch/);
ClinicalTrials.gov (http://clinicaltrials.gov/);
Current Controlled Trials (http://www.controlled‐trials.com/).

Author searches

The review authors searched the following Chinese databases for published and unpublished studies:

Chinese BioMedical Database (1980 to 28 February 2014) (http://www.imicams.ac.cn/);
Traditional Chinese Medical Database System (1984 to 28 February 2014) (http://cowork.cintcm.com/engine/windex.jsp);
China National Knowledge Infrastructure (1994 to 28 February 2014) (http://acad.cnki.net/kns55/default.aspx);
Chinese VIP Information (1989 to 28 February 2014) (http://cstj.cqvip.com/);
Chinese Medical Current Contents (1994 to 28 February 2014) (http://210.34.66.109/esource/showdb.jsp?ID=6);
Chinese Academic Conference Papers Database and Chinese Dissertation Database (1994 to February 2014) (http://acad.cnki.net/kns55/brief/result.aspx?dbPrefix=CMFD);
China Medical Academic Conference (1994 to 28 February 2014) (http://acad.cnki.net/kns55/Navigator.aspx?ID=CPFD);
WangFang database (1987 to 28 February 2014) (http://www.wanfangdata.com.cn/).

The search strategies which were used are listed in Appendix 5 to Appendix 12.

Searching other resources

We searched the bibliographies of all relevant publications for further studies. One review author (Zheng GH) contacted authors and experts in the field of surgical nursing (Guan FG, Zhao RH) for any information about unpublished studies.

Data collection and analysis

Selection of studies

Two review authors (Chen HY and Chu JF) independently screened the abstract, title, or both sections of every record retrieved according to the inclusion criteria. The full text of literature with any unclear information in the title or abstract was retrieved for clarification. We investigated all potentially relevant articles as full text. We selected the most complete publication or pooled the data if there were two or more publications relating to one trial. Any disagreements were resolved by discussion.

Data extraction and management

Mei LJ and Yang L independently extracted data using a data extraction form and the results were checked for accuracy by a third author (Zheng GH). Disagreements were resolved by discussion. In the case of duplicate publications and companion papers of a primary paper, we maximised the yield of information by simultaneous evaluation of all available data. The information that was extracted from each included report included the following.

1. Basic information: study design, title, authors, publication status.

2. Participant characteristics: mean age, proportion of each gender, sample number.

3. Methodological description: method of randomisation, method of concealment of allocation, exclusions post‐randomisation, blinding, losses to follow up.

4. Intervention characteristics: type of intervention (prevention or treatment), routes of administration, period of treatment, duration of follow up.

5. Outcomes: primary and secondary outcomes, safety outcomes.

6. Other: infusion given, site of vein, type of catheter, risk factors for phlebitis.

Assessment of risk of bias in included studies

Two review authors (Zheng GH, Yang L) independently assessed the risk of bias of each trial by using the tool developed by The Cochrane Collaboration (Higgins 2011) and the following criteria were used. Disagreements were resolved by consensus.

1. Was the allocation sequence randomly generated?

2. Was the treatment allocation adequately concealed?

3. Was knowledge of the allocated interventions adequately prevented during the study?

4. Were incomplete outcome data adequately addressed?

5. Are reports of the study free of suggestion of selective outcome reporting?

6. Was the study apparently free of other problems that could put it at a high risk of bias?

Measures of treatment effect

Dichotomous outcomes

We expressed dichotomous data (for example the rate of resolution of phlebitis, incidence of phlebitis) as the risk ratio (RR) with its 95% confidence interval (CI). We pooled data using the fixed‐effect model or the random‐effects model when heterogeneity among studies was obvious. We did not calculate the risk difference (RD) or include the number needed to treat (NNT) in the analysis because this was not feasible with the data available. We did not perform the evaluation of adverse events because the data were unavailable in the included studies. If adverse events data are available in future updates we will calculate the number needed to harm (NNH).

Continuous data

We analysed continuous outcomes by using mean differences (MD) with 95% CIs.

Unit of analysis issues

The primary analysis was based on per individual randomised. Cross‐over trials and cluster‐randomised trials were not found for this review. For trials with more than two intervention groups, we combined groups to create a single pair‐wise comparison (Higgins 2011a).

Dealing with missing data

We attempted to obtain relevant missing data from the authors of the included trials, if feasible, and carefully performed evaluations of important numerical data such as the number of screened and randomised participants as well as the intention‐to‐treat (ITT) and per protocol (PP) populations.

Assessment of heterogeneity

We tested the heterogeneity between the included trials by examining the I² statistic, quantifying inconsistency across studies, to assess the impact of heterogeneity on the meta‐analysis (Higgins 2003). We analysed the study components such as the participants, diseases, interventions, comparisons and outcomes if obvious heterogeneity (I² > 75%) was found, and explained the source of heterogeneity by subgroup analysis and sensitivity analysis. We used a random‐effects model meta‐analysis as an overall summary if the heterogeneity among trials was considered acceptable (85% > I² > 50%). If heterogeneity was substantial (I² > 85%) we did not pool the data and produced a narrative qualitative summary.

Assessment of reporting biases

We used funnel plots to assess publication bias when a sufficient number of studies were included (at least 10 studies) as recommended by the Cochrane Handbook for Systematic Reviews of Interventions 5.1.0 (Higgins 2011; Sterne 2001).

Data synthesis

For the predefined outcomes which were available, measured similarly, and of sufficient quality we statistically summarised the pooled effects and performed statistical analyses according to the statistical guidelines referenced in the current version of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a).

We performed a meta‐analysis of the data using a fixed‐effect model prior to a random‐effects model only if the test of heterogeneity was not significant. However, if heterogeneity was substantial (I² > 85%), we did not perform a meta‐analysis at all and a narrative, qualitative summary was carried out.

Subgroup analysis and investigation of heterogeneity

We carried out the following subgroup analyses to explore possible sources of heterogeneity, for primary outcomes only.

1. Duration of treatment (such as Aloe vera used for three, five or seven days, etc).

2. Types of Aloe vera products (such as a thin slice or the juice of Aloe vera and Aloe‐derived products).

Sensitivity analysis

Where possible, we performed the following sensitivity analyses to explore the effect of potential biases.

1. Repeating the analysis excluding unpublished studies.

2. Repeating the analysis taking account of study quality.

3. Reanalysing the data using different statistical methods.

4. Repeating the analysis excluding qRCTs.

Results

Description of studies

For a detailed description of the studies, see Characteristics of included studies; Characteristics of excluded studies.

Results of the search

See Figure 1.

Figure 1

Study flow diagram.

Included studies

A total of 43 studies (35 RCTs and eight qRCTs) with 7465 participants were included in the review (Cao 2008; Chen 2009; Chen 2010; Chen 2012; Deng 2010; Deng 2012; Dong 2001; Dong 2008; Gao 2007; Gao 2012; Hou 2010; Hu 2009; Ji 2007; Jin 2010; Li 2003; Li 2006; Li 2009; Li 2011; Liang 2004; Liu 2003; Liu 2006; Liu 2012; Liu 2012b; Lu 2004; Pan 2008; Peng 2009; Ren 2008; Tan 2002; Tang 2011; Wang 2006; Wang 2008; Wang 2010; Wang 2010b; Wu 2009; Xiao 2008; Yang 2008; Yao 2009; Yu 2006; Zhang 2010; Zhang 2013; Zheng 2010; Zhong 2011; Zhou 2006).

The number of participants in each trial ranged from 40 (Chen 2009) to 3000 (Hu 2009). The age of participants ranged from 18 to 75 years old except for one study where the age of participants ranged from one month to three years old (Gao 2007). All included studies were conducted in China, and were performed in the hospital setting.

Various formulations of Aloe vera or aloe‐derived products were used in the included studies. A slice of fresh aloe 5 cm to 15 cm long and 3 cm wide was most commonly used and reported on in 31 included studies. Six studies used fresh aloe leaf divided into pieces 15 cm to 20 cm long and 5 cm to 10 cm wide (Hou 2010; Hu 2009; Pan 2008; Wang 2010b; Xiao 2008; Zhang 2010). Other aloe vera products used were the juice of fresh aloe leaf in three included studies (Deng 2010; Jin 2010; Ren 2008), aloe glue in two studies (Chen 2010; Li 2011), and dry aloe linimentum in one study (Lu 2004).

Twenty‐two studies with 5546 participants involved the use of Aloe vera for prevention of phlebitis (Cao 2008; Chen 2012; Dong 2008; Hou 2010; Hu 2009; Ji 2007; Li 2011; Liang 2004; Liu 2003; Liu 2006; Liu 2012b; Pan 2008; Peng 2009; Ren 2008; Tan 2002; Wang 2006; Wu 2009; Xiao 2008; Yao 2009; Yu 2006; Zhang 2010; Zheng 2010). Of these 22 studies, 13 trials with 4272 participants compared Aloe vera alone in the treatment group with no treatment in the control group (Cao 2008; Dong 2008; Hu 2009; Li 2011; Liu 2006; Liu 2012b; Pan 2008; Peng 2009; Tan 2002; Xiao 2008; Yao 2009; Yu 2006; Zheng 2010), and two trials with 248 participants compared Aloe vera alone with 50% MgSO₄ (Ren 2008; Zhang 2010). Two trials with 200 participants compared Aloe vera versus 33% MgSO₄ (Chen 2012; Hou 2010). One trial with 66 participants compared Aloe vera alone with potato slices (Wu 2009). One trial with 150 participants compared Aloe vera alone with 75% alcohol (Liang 2004). One trial with 147 participants compared aloe plus an ice bag compress with an ice bag compress plus 75% MgSO₄ (Liu 2003). One study was a three‐arm trial with 320 participants comparing Aloe vera versus potato slice versus no treatment (Wang 2006). The remaining trial with 142 participants compared Aloe vera plus saline (NaCl₂) plus dexamethasone with NaCl₂ plus dexamethasone in the control group (Ji 2007).

The therapeutic effect of Aloe vera for phlebitis was evaluated in 21 studies involving 1919 participants (Chen 2009; Chen 2010; Deng 2010; Deng 2012; Dong 2001; Gao 2007; Gao 2012; Jin 2010; Li 2003; Li 2006; Li 2009; Liu 2012; Lu 2004; Tang 2011; Wang 2008; Wang 2010; Wang 2010b; Yang 2008; Zhang 2013; Zhong 2011; Zhou 2006). Of these, 10 studies with 880 participants compared aloe alone in the treatment group with 50% MgSO₄ in the control group (Chen 2010; Deng 2010; Deng 2012; Gao 2007; Li 2003; Li 2009; Liu 2012; Tang 2011; Wang 2010; Yang 2008), and three studies with 422 participants compared Aloe vera alone with 33% MgSO₄ (Dong 2001; Gao 2012; Li 2006). The remaining eight studies compared Aloe vera alone or Aloe vera plus other treatment, such as sulphanilamide pyrimidine, potato, sulphonic acid mucopolysaccharide cream, in the treatment group versus the same 'other treatment' in the control group (Chen 2009; Jin 2010; Lu 2004; Wang 2008; Wang 2010b; Zhang 2013; Zhong 2011; Zhou 2006).

Excluded studies

In total 60 studies were excluded on the basis of their full text versions. Reasons for exclusion were as follows: self‐controlled clinical trial (Zou 2004), an incorrect control used (the non‐Aloe vera treatment used in the treatment group but not used in the control group) in 50 studies (Cao 2007; Dai 2007; Deng 2001; Deng 2009; Gong 2004; Gu 2013; Guo 2009; Hu 2011; Huang 2005, Huang 2009; Jiang 2007; Li 2003b; Li 2006b; Li 2010; Li 2011b; Liang 2011; Lin 2009; Lu 2004b; Lu 2007; Lu 2008; Lu 2010; Lu 2010b; Lu 2011; Ma 2008; Ma 2009; Mo 2008; Peng 2002; Shao 2008; Shen 2013; Shi 2013; Su 2011; Tang 2008; Tang 2008b; Wang 2004;Wang 2010c;Wang 2011;Wang 2012; Wang 2013; Wei 2012; Wu 2010;Xu 2007;Xu 2008;Yan 2012; Yang 2010;Ye 2008; Ye 2011;Zhang 2004; Zhang 2008; Zhu 2010; Zhuang 2008), and an unclear intervention duration in nine studies (Chen 2007; Huang 2013; Lei 2012; Liu 2009; Zhang 2003; Zhang 2005; Zhang 2006; Zhang 2011; Zhou 2001). Although we were able to contact one of the authors, we did not obtain more information to judge the duration of the intervention.

Risk of bias in included studies

Of the 43 included studies, all included studies were judged as at high risk of bias because one or more main aspects of the bias assessment was labelled high or unclear risk of bias. We tried to contact the original study authors but we received no response from most of them. See Figure 2 and Figure 3.

Figure 2

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Figure 3

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

Four studies with adequate sequence generation used random digit tables or computer random digit generators (Cao 2008; Jin 2010; Tang 2011; Wang 2010b). Eight studies were of qRCT design and judged as at high risk of bias because of alternate allocation using the participant' birthdays or admission sequences (Deng 2010; Gao 2007; Hou 2010; Hu 2009; Liu 2006; Wang 2006; Xiao 2008; Zhang 2010). In the other studies the method was unclear because of a lack of useful information reported in the study papers.

Adequate allocation concealment, by sealed opaque envelopes, was used in one study (Cao 2008). The eight qRCTs were judged to be at high risk of bias for allocation concealment as described above (Deng 2010; Gao 2007; Hou 2010; Hu 2009; Liu 2006; Wang 2006; Xiao 2008; Zhang 2010). The remainder of the studies were judged to be at unclear risk of bias because of insufficient reporting.

Blinding

Blinding of participants and nurses was not possible due to the external application of Aloe vera. Therefore, we assessed blinding as at high risk of bias in all included studies. No included studies reported the blinding of outcome assessors, and we judged this risk of bias to be unclear because of a lack of sufficient information reported in the papers.

Incomplete outcome data

The flow of participants was not reported in any of the included studies so we were uncertain whether or not the reported outcome data were appropriate. One study reported that the withdrawal and attrition rates between the treatment and control groups were similar (Cao 2008) and this study was therefore judged to be at low risk of bias. For the remaining 42 studies it was not clear how many were randomised and how many withdrawn following randomisation.

Selective reporting

No study protocols were available for any of the included studies. Therefore, it was difficult to judge whether selective reporting bias existed in the studies. Only for Cao 2008 it was clear that the expected outcomes were reported. We judged the remaining studies to be at unclear risk of bias.

Other potential sources of bias

The baseline information were considered to be similar between the comparison groups in 23 (Cao 2008; Chen 2009; Chen 2010; Deng 2010; Dong 2008; Hou 2010; Hu 2009; Li 2009; Li 2011; Liang 2004; Liu 2006; Lu 2004; Peng 2009; Ren 2008; Tan 2002; Wang 2006; Wang 2008; Wang 2010b; Yao 2009; Yu 2006; Zheng 2010; Zhong 2011; Zhou 2006) of the included studies. The remaining 20 studies (Chen 2012; Deng 2012; Dong 2001; Gao 2007; Gao 2012; Ji 2007; Jin 2010; Li 2003; Li 2006; Liu 2003; Liu 2012; Liu 2012b; Pan 2008; Tang 2011; Wang 2010; Wu 2009; Xiao 2008; Yang 2008; Zhang 2010; Zhang 2013) were judged as at unclear risk of bias because they lacked a detailed description of the baseline information.