Home > Health & Research > Health Education Campaigns & Programs > Cochrane Neonatal Review > Pentoxifylline for treatment of sepsis and necrotizing enterocolitis in neonates

Pentoxifylline for treatment of sepsis and necrotizing enterocolitis in neonates

Skip sharing on social media links
Share this:

Authors

Khalid N Haque1, Pammi Mohan2

Background - Methods - Results - Characteristics of Included Studies - References - Data Tables & Graphs


1Division of Neonatology, Department of Child Health, Queen Mary's Hospital for Children, Wrythe Lane, Carshalton, UK [top]
2Pediatrics, Section of Neonatology, Baylor College of Medicine, Houston, USA

Contact person

Pammi Mohan

Assistant Professor
Pediatrics, Section of Neonatology
Baylor College of Medicine
6621, Fannin, MC.WT 6-104
Houston Texas 77030
USA

E-mail: mohanv@bcm.tmc.edu

Dates

Assessed as Up-to-date: 11 September 2007
Date of Search: 30 May 2007
Next Stage Expected: 11 September 2009
Protocol First Published: Issue 2, 2003
Review First Published: Issue 4, 2003
Last Citation Issue: Not specified

What's new

Date / Event Description
11 September 2007
Updated

This updates the review "Pentoxifylline for neonatal sepsis" published in the Cochrane Database of Systematic Reviews, Issue 2, 2003 (Haque 2003).

The updated search did not identify any new trials. Two ongoing trials using pentoxifylline in the treatment of necrotizing enterocolitis were identified.

The title and the review has been modified to include pentoxifylline treatment for necrotizing enterocolitis, in view of emerging evidence for potential benefits of the use of pentoxifylline for this condition.

13 February 2008
Amended

Converted to new review format.

History

Date / Event Description
28 January 2003
New citation: conclusions changed

Substantive amendment

Abstract

Background

Mortality and morbidity due to neonatal sepsis and necrotizing enterocolitis (NEC) is high despite the use of potent antimicrobial agents. Agents that modulate inflammation may improve outcome. Pentoxifylline, a phosphodiesterase inhibitor, is one such agent.

Objectives

The primary objectives were to assess the. Effect on mortality and the safety of intravenous pentoxifylline as an adjunct to antibiotic therapy in neonates with

  1. suspected or confirmed sepsis
  2. NEC.

Search strategy

The Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 2, 2007), MEDLINE, EMBASE and CINAHL, Science Citation Index for articles referencing Lauterbach 1996, proceedings of the Pediatric Academic Societies (1980 - 2006), BIOSIS (1992-2007), conference proceedings (1992 - 2007), ongoing trials and reference lists of identified RCTs were searched in May 2007.

Selection criteria

Randomised or quasi-randomised trials assessing the efficacy of pentoxifylline as an adjunct to antibiotics for treatment of suspected or confirmed sepsis or NEC in neonates and reporting at least one of the following outcomes: all cause mortality during initial hospital stay, neurological development at two years of age or later, length of hospital stay, duration of ventilation via endotracheal intubation, chronic lung disease in survivors, periventricular leukomalacia, necrotising enterocolitis, or adverse events were eligible.

Data collection and analysis

Two reviewers independently abstracted information for the outcomes of interest. Typical Relative Risk (RR) and Risk Difference (RD) with 95% confidence intervals (CI) using fixed effects model are reported for dichotomous outcomes. NNT was calculated for outcomes for which there was a statistically significant reduction in RD.

Results

Two randomized controlled trials that enrolled a total of 140 preterm (< 36 weeks gestation) neonates with suspected late onset (> 7 days) sepsis were identified. Outcomes were reported for only 107 randomised patients with confirmed sepsis. A reduction was noted in 'all cause mortality during hospital stay' following pentoxifylline treatment [typical RR 0.14 (95% CI 0.03, 0.76), typical RD -0.16 (95% CI -0.27, - 0.04), NNT 6 (95% CI 4, 25)]. No adverse effects due to pentoxifylline or other outcomes of interest were reported. No completed trial of treatment with pentoxifylline for NEC was identified.

Authors' conclusions

Current evidence from two small studies suggests that the use of pentoxifylline as an adjunct to antibiotics in neonatal sepsis reduces mortality without any adverse effects. Researchers are encouraged to undertake large well-designed trials to confirm or refute the effectiveness of pentoxifylline in reducing mortality and adverse outcomes in neonates with suspected or confirmed neonatal sepsis and NEC.

Plain language summary

Pentoxifylline for treatment of sepsis and necrotizing enterocolitis in neonates

Pentoxifylline added to antibiotic treatment may reduce mortality from sepsis in newborn babies, but more research is needed. Sepsis is a bacterial or fungal infection of the bloodstream. Necrotizing enterocolitis (NEC) is a condition in premature babies where the gastrointestinal tract is damaged. NEC is often associated with infection. Sepsis and NEC may cause death and adversely affect development. Modulating the body's response to infection (inflammation) may reduce death and tissue injury after sepsis and NEC. Pentoxifylline is a drug that affects inflammation. This review found some evidence that pentoxifylline in combination with antibiotics reduces deaths from sepsis in the newborn. No adverse effects due to pentoxifylline were identified. Although two ongoing studies were found, there were no completed studies looking at treatment of NEC with pentoxifylline. More research is needed on pentoxifylline and other anti-inflammatory drugs that might be useful the treatment of sepsis and NEC in the newborn.

[top]

Background

Neonatal sepsis is the most common cause for neonatal deaths worldwide (Lawn 2006). The incidence of neonatal sepsis in the developed world is reported to be between 0.6 to 1.2% (Philip 1994; Stoll 2004a). In the developing world, the incidence of sepsis is significantly higher (20 to 40 % of all live births) (Bhutta 1997). The incidence of neonatal sepsis is inversely proportional to gestational age and birth weight (Kaufman 2004). Early onset sepsis (sepsis in infants < 72 hrs of life) occurs in 1.5 to 1.9% of very low birthweight infants (VLBW, birthweight 401 - 1500 gm) (Stoll 2005). In a cohort of 6956 VLBW infants admitted to the NICHD Neonatal Network hospitals for the period 1998-2000, 21% had one or more blood culture proven late onset sepsis (onset after 72 hours of life) (Stoll 2002). Mortality of infants with late onset sepsis was 18% (36% for those infected with gram negative organisms) and morbidities (including patent ductus arteriosus, prolonged ventilation, prolonged need for intravascular access, bronchopulmonary dysplasia, necrotizing enterocolitis and length of hospital stay) were significantly higher in infected infants. Sepsis significantly affects long-term neurodevelopmental outcomes, either by direct infection of the central nervous system or as a result of inflammatory injury. In a large cohort study of 6093 extremely low birth weight infants (ELBW, birth weight < 1000 g), infected infants had a significantly higher incidence of adverse neurodevelopmental outcomes at follow-up, characterized by cerebral palsy, low Bayley's scores of infant development and vision impairment when compared with uninfected infants (Stoll 2004b).

Necrotizing enterocolitis (NEC) occurs in about 1 to 5% of infants admitted to the neonatal intensive care unit. The most consistent risk factors are prematurity and low birthweight (Lin 2006). The pathogenesis of NEC is not entirely clear. Gastrointestinal immaturity, enteral feeding (especially formula feeding), presence of bacteria and inflammation all play a part in the development of NEC (Lin 2006). TNF-α and platelet activating factor are the most important among pro-inflammatory cytokines that have been implicated in the development of NEC (Caplan 1990; Caplan 1990a). The pivotal role of TNF-α in NEC is supported by the fact that, in an animal model, monoclonal antibody to TNF-α reduced incidence of NEC from 80% - 17% (Halpern 2006). Mortality from NEC is high (15 - 30%) and 20 - 40% of infants with NEC undergo surgery. Infants who had NEC have delayed neurodevelopmental outcomes at 18 - 22 months corrected age (Lin 2006, Stoll 2004b).

Mortality and morbidity due to sepsis and NEC remain high despite the use of potent antimicrobial agents (Stoll 2002; Stoll 2005). Increased use of antimicrobials has led to a global emergence of antibiotic resistance (Levy 1998). Adjunct therapies may be important in increasing the efficacy of antimicrobial agents. Excessive or uncontrolled inflammatory response may be responsible for the multi-organ dysfunction and systemic inflammatory response (SIRS) seen in sepsis. The balance of pro and anti-inflammatory cytokines may determine the severity and the ultimate outcome in sepsis syndromes and NEC (Edelson 1999; Ng 2003; Harris 2005). Recent evidence has also shown that inflammation plays an important role in cerebral (Adams-Chapman 2006) and pulmonary injury (Speer 1999), especially in the preterm neonate. When used in conjunction with antibiotics, immunomodulating agents may help to re-establish the balance between pro and anti-inflammatory responses and may influence clinical outcome in neonatal sepsis and NEC.

Pentoxifylline (PTX), a xanthine derivative, is a phosphodiesterase inhibitor that suppresses TNF-α production by adenylcyclase activation and increased cellular cyclic AMP concentration. PTX has attracted increased interest since the discovery that inhibition of tumor necrosis factor gene transcription reduces mortality from sepsis. TNF-α increases peroxidation of arachidonic acid, activates polymorphonuclear leukocytes, increases eicosanoids and increases its own production and thereby amplifies the inflammatory response (Jaattela 1991; Vilcek 1991). Inhibition of TNF-α production by PTX negates this response and thereby may improve outcome. PTX also has beneficial effects on endothelial cell function and coagulation in sepsis (Wang 1996; Boldt 1996).

Pentoxifylline has been shown to have beneficial effects in humans and animal models of sepsis and necrotizing enterocolitis. In sepsis, PTX has been shown to improve haemodynamics (including renal blood flow) and to prevent transition from a hyperdynamic to a hypodynamic response (Zeni 1996; Bacher 1997; Krysztopik 1996; Yang 1999). Inflammatory lung injury after endotoxemia is also ameliorated by PTX (Michetti 2003). In adults and neonates, PTX has been shown to decrease serum levels TNF-α, IL-1, and IL-10 but not IL-6 or IL-8 (Zeni 1996; Bienvenu 1995). In a rat model of NEC, PTX reduces the incidence and severity of NEC (Travadi 2006).

To date, no significant adverse effects have been reported in either animal or human studies. The potential beneficial effects make pentoxifylline a promising agent for the treatment of sepsis and NEC in neonates. This systematic review evaluates the effect of pentoxifylline as an adjunct to antibiotics for the treatment of sepsis and NEC in neonates.

Objectives

Primary objectives:

  1. To assess the effect of intravenous pentoxifylline as an adjunct to antibiotic therapy on mortality and morbidity in neonates with suspected or confirmed sepsis
  2. To assess the effect of intravenous pentoxifylline as an adjunct to antibiotic therapy on mortality and morbidity in neonates with NEC

Separate comparisons will be made for pentoxifylline treatment for neonatal sepsis and NEC.

Secondary objectives:

  1. For treatment of neonatal sepsis:
    1. To determine the effect of pentoxifylline for treatment of neonatal sepsis on adverse neurological outcome at two years of age or later
    2. To determine the effect of pentoxifylline for treatment of neonatal sepsis on the length of hospital stay in survivors to discharge and to determine adverse effects attributable
    3. to pentoxifylline (e.g. gastrointestinal disturbance, increasing gastric residue, feeding intolerance, thrombocytopenia, cholestatic jaundice)
    4. To determine the effect of pentoxifylline for treatment of neonatal sepsis in the following subgroups of neonates; term or preterm infants with suspected or confirmed
    5. sepsis, gram negative sepsis, fungal sepsis, early onset sepsis (within first 72 hrs of life) or late onset sepsis (> 72 hrs of life)
    6. To determine the effect of pentoxifylline for treatment of neonatal sepsis on the duration of ventilation through endotracheal tube, development of chronic lung disease and necrotising enterocolitis
  2. For treatment of NEC:
    1. To determine the effect of pentoxifylline for treatment of NEC on adverse neurological outcome at two years of age or later
    2. To determine the effect of pentoxifylline for treatment of NEC on the length of hospital stay in survivors to discharge and to determine adverse effects attributable to pentoxifylline (e.g. gastrointestinal disturbance, increasing gastric residue, feeding intolerance, thrombocytopenia, cholestatic jaundice)
    3. To determine the effect of pentoxifylline for treatment of NEC in the following subgroups of neonates; term or preterm infants
    4. To determine the effect of pentoxifylline for treatment of NEC on the duration of ventilation through endotracheal tube and development of chronic lung disease

[top]

Methods

Criteria for considering studies for this review

Types of studies

Randomised or quasi-randomised controlled trials

Types of participants

Neonates (< 28 days old, at any gestational age or birthweight) with confirmed or suspected sepsis or neonates with NEC [Bell's Stage II or III (Bell 1978)] on antibiotics

Confirmed sepsis was defined as clinical signs and symptoms consistent with infection and microbiologically proven, with a positive blood culture, CSF culture, urine culture (obtained by a suprapubic tap) or culture from a normally sterile site (e.g. pleural fluid, peritoneal fluid or autopsy specimens) for bacteria or fungi

Suspected sepsis was defined as clinical signs and symptoms consistent with sepsis without isolation of a causative organism

NEC was defined as an acute gastrointestinal disorder that manifests clinically with systemic signs (temperature instability, apnea, bradycardia, lethargy, hypotension, metabolic acidosis, hyponatremia, thrombocytopenia, disseminated intravascular coagulation), intestinal signs (feed intolerance, GI bleeding, abdominal tenderness, abdominal wall cellulitis, abdominal distension), radiological features (non-specific intestinal dilation and ileus in Stage I, by pneumatosis intestinalis and air in the portal tree in Stage II, or pneumoperitoneum in Stage III) and pathologically by intestinal necrosis (Bell 1978; Walsh 1986).

Planned subgroup analyses included

  1. Gestational age
    1. Preterm neonates (born before 37 completed weeks gestation)
    2. Term infants (born at or after 37 completed weeks of gestation)
  2. Time of onset of sepsis
    1. Early onset sepsis (sepsis in the first 72 hrs of life)
    2. Late onset sepsis (sepsis after the first 72 hrs of life)
  3. Suspected or confirmed sepsis
    1. Neonates with suspected sepsis (clinical signs and symptoms consistent with sepsis without isolation of causative organism treated with antibiotics).
    2. Neonates with confirmed sepsis
    3. Neonates with confirmed gram negative sepsis
    4. Neonates with confirmed fungal sepsis

Types of interventions

Intravenous pentoxifylline at any dosage or duration used as adjunct to antibiotics to treat suspected or confirmed neonatal sepsis or NEC, compared with placebo, or no intervention, or other adjuncts to antibiotics.

Types of outcome measures

Primary outcomes:
  • All cause mortality during hospital stay
Secondary outcomes:
  • Neurological outcome at two years of age or more (neurodevelopmental outcome assessed by a validated test)
  • Chronic lung disease (CLD) in survivors (CLD defined as oxygen requirement at 36 weeks postmenstrual age)
  • Adverse outcomes directly attributable to pentoxifylline: thrombocytopenia, (platelet count less than 100, 000 x 109/L) increased gastric residue (gastric aspirate greater than 10% of oral feed), vomiting, cholestatic jaundice requiring therapy
  • Periventricular leukomalacia (defined as necrosis of white matter in a characteristic distribution, i.e. in the white matter dorsal and lateral to the external angles of lateral ventricles involving particularly the centrum semi ovale, optic and acoustic radiations and diagnosed by neuroimaging) (Volpe 1995)
  • Duration of assisted ventilation through an endotracheal tube in days.
  • Length of hospital stay in days for survivors to discharge
  • Outcome for neonates with sepsis only: Necrotising enterocolitis (NEC) (definite NEC and perforated NEC, Bell's stage II or III) (Bell 1978)

Search methods for identification of studies

Please refer to Cochrane Neonatal Review Group's search strategy.

The following databases were searched again in May 2007 and relevant trials in any language were identified through:

  1. The Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 2, 2007)
  2. Electronic journal reference databases: MEDLINE (1966 - present) and PREMEDLINE, EMBASE (1980 - present), CINAHL (1982 - present)
  3. Abstract of conferences - proceedings of Pediatric Academic Societies (American Pediatric Society, Society for Pediatric Research) and European Society for Paediatric Research were searched from 1990 in the journal Pediatric Research and 'Abstracts online' (2000-2006).
  4. Biological abstracts in the database BIOSIS and conference abstracts from 'Proceedings First' from 1992-2007 were searched but none were identified.
  5. Science citation index was searched for all articles that cited ' Lauterbach 1996'.
  6. Ongoing trials were searched with the search engines provided at the web sites www.clinicaltrials.gov and www.controlled-trials.com. One ongoing trial (Hammerman 2005) was identified with the search word 'pentoxifylline'.
  7. Authors who published in this field were contacted for possible unpublished studies and one ongoing study was identified (Patole 2006).
  8. Additional searches were made from the reference lists of identified clinical trials and in the reviewer's personal files.

MEDLINE and PREMEDLINE search strategy. This was adapted to suit EMBASE, CINAHL and the Cochrane Controlled Trials Register.

  1. explode 'sepsis' [all subheadings in MIME, MJME]
  2. sepsis or septicemia
  3. septic
  4. NEC
  5. 5 'necrotizing enterocolitis'
  6. # 1 or # 2 or # 3 or # 4 or # 5
  7. explode 'infant - newborn' [all subheadings in MIME, MJME]
  8. Neonat*
  9. Newborn*
  10. # 7 or # 8 or # 9
  11. # 6 and # 10
  12. "Pentoxifylline' [all subheadings on MIME, MJME]
  13. pentoxif*
  14. Trental or Torental
  15. PTF or PTX or PTX F
  16. agapurin or BL-191 or oxpentif*
  17. # 12 or # 13 or # 14 or # 15 or # 16
  18. # 11 and # 17

No language restriction was applied

Data collection and analysis

The titles and the abstracts of studies identified by the search strategy were assessed independently for eligibility for inclusion in this review by the two reviewer authors. If this could not be done reliably by title and abstract, then the full text version was obtained for assessment. Any differences were resolved by mutual discussion. Full text versions of all eligible studies were obtained for quality assessment. Forms were designed for trial inclusion/exclusion, data extraction and for requesting additional information from authors of the original reports. Data extraction was done independently by the authors using specifically designed paper forms and compared for any differences which were then resolved by discussion.

Assessment of the quality of studies:

The standardized review methods of the Cochrane Neonatal Review Group (CNRG) were used to assess the methodological quality of the studies. An assessment of the concealment of allocation in the included studies was done by the two reviewers independently using the standard criteria developed by the CNRG.

  1. Adequate allocation concealment
  2. Uncertainty about whether the allocation was adequately concealed
  3. Inadequate allocation concealment

In addition, blinding of intervention, blinding of outcome assessment and completeness of follow-up of all randomised infants is reported. Each of these criteria were then rated as yes, no or unsure.

Analysis

Statistical analyses were performed according to the recommendations of the CNRG. Analyses were done for all infants, and for the subgroups defined under 'Criteria for considering studies for this review '. All infants randomised were analysed on 'an intention to treat basis' irrespective of whether or not they survived to receive their allocated treatment completely. Treatment effects in the individual trials were analysed. Heterogeneity of treatment effects between trials was assessed to check the appropriateness of pooling data and performing meta-analysis. The statistical package (RevMan 4.1) provided by the Cochrane Collaboration was used. Relative risk (RR) and risk difference (RD) with 95% confidence interval (CI) are reported for dichotomous outcomes. If there was a statistically significant reduction in RD then the number needed to treat (NNT) was calculated. A fixed effects model was used for meta-analysis.

[top]

Results

Description of studies

Details of the included studies are provided in the table "Characteristics of Included Studies".

Two studies (Lauterbach 1996; Lauterbach 1999) met the inclusion criteria.

Lauterbach 1996 randomised 40 preterm infants (< 36 weeks gestation) with suspected late onset sepsis to receive either pentoxifylline or placebo as adjunct to antibiotics. In 4/20 infants in the treatment group and 7/20 infants in the placebo group sepsis was not confirmed, and these infants were excluded from analysis in the report of this trial. Thus, outcomes were reported in only the 29 infants with confirmed sepsis (16 in treatment and 13 in placebo group). Of these, confirmed gram negative sepsis occurred in 10 infants in the pentoxifylline group and in 10 infants in the placebo group. The following outcomes were reported: 1) plasma TNF-α levels, 2) mortality during hospital stay, and 3) adverse effects.

Lauterbach 1999 randomised 100 preterm infants (< 36 weeks gestation) with suspected late onset sepsis to receive either pentoxifylline or placebo as adjunct to antibiotics. 10/50 infants in the treatment group and 12/50 in the placebo group were excluded from analysis in the report of this trial as sepsis was not confirmed in these infants. Thus, outcomes were reported for only the 78 infants with confirmed sepsis (40 in the pentoxifylline and 38 in the placebo group). Of these, confirmed gram negative sepsis occurred in 15 infants in the pentoxifylline group and 14 infants in the placebo group. The following outcomes were reported: 1) plasma levels of TNF-α, IL-6 and IL-1, 2) mortality during hospital stay and 3) adverse effects.

Excluded studies:

Lauterbach 1994: 17 preterm infants with sepsis were given pentoxifylline and compared with a historical control group of 13 septic infants who did not receive pentoxifylline. Mortality and adverse effects were reported. The study was excluded as it was neither a randomised nor a quasi-randomised study.

Ongoing studies:

Hammerman 2005 and coworkers plan to randomise 50 premature neonates with a birthweight less than 1750 g and with clinical and radiological diagnosis of NEC to pentoxifylline (5 mg/kg/hr for 6 hr for 6 days) or an equal volume of 1/2 normal saline. Infants with abdominal perforation will be excluded. This study started in January 2005 and scheduled to conclude in March 2009.

In a pilot study, Patole 2006 and coworkers plan to randomise 80 preterm (< 32 weeks gestation) neonates with Stage II or III NEC. Enrolled neonates will be randomised to intravenous pentoxifylline or an equal volume of placebo at 5 mg/kg/hour for 12 hours a day (60 mg/kg/day) for two consecutive days, followed by infusion for six hours a day (30 mg/kg/day) for the next four consecutive days. The primary outcome is to assess the efficacy and safety of pentoxifylline in preventing the progression of NEC and or death. Secondary outcomes are reduction in plasma TNF-α levels, duration of hospital stay, duration of TPN support and time to full enteral feeds. This study is scheduled to be conclude in 2009 or 2010.

Risk of bias in included studies

See: Table of included studies.

Lauterbach 1996: Single centre randomised placebo controlled trial. No details of randomisation are available, thus blinding of randomisation is unclear. Physicians were blinded to the intervention. Blinding of outcome assessment unclear. 11/40 were excluded from analysis as they did not have confirmed sepsis.

Lauterbach 1999: A two centre randomised placebo controlled trial. Randomisation was done using computer generated random numbers centrally. There was concealment of randomisation. Physicians were blinded to the intervention and outcome. 22/100 were excluded from analysis as they did not have confirmed sepsis.

In both the above trials, information has been requested from the author regarding the outcomes of infants with suspected late onset sepsis who were excluded from the analysis.

Effects of interventions

I. PENTOXIFYLLINE VERSUS PLACEBO FOR TREATMENT OF NEONATAL SEPSIS (COMPARISON 01):

Two randomised controlled trails where pentoxifylline was used for the treatment of neonatal sepsis were eligible for inclusion in the review.

A total of 140 preterm infants with suspected neonatal sepsis (Lauterbach 1996, Lauterbach 1999) have been enrolled in RCTs to evaluate the effect of pentoxifylline on important clinical outcomes. A total of 107 neonates with subsequently confirmed sepsis were analysed in the reports of these trials.

Primary outcomes
1. All cause mortality during hospital stay (Outcome 01.01):

All infants with confirmed sepsis (2 trials, 107 infants, 10 events) (Outcome 01.01.01):

There is a significant reduction in all cause mortality during hospital stay in neonates with confirmed sepsis who had pentoxifylline as an adjunct to antibiotics compared to neonates with sepsis who had placebo [typical RR 0.14 (95% CI 0.03, 0.76), typical RD -0.16 (95% CI -0.27, -0.04), NNT 6 (95%CI 4, 25)]. There was no significant heterogeneity in this result for the two trials.

Subgroup analyses:

Confirmed gram negative sepsis (2 trials, 49 infants, 8 events) (Outcome 01.01.02):

There is a reduction of borderline statistical significance in all cause mortality during hospital stay in neonates with confirmed gram negative sepsis who had pentoxifylline as an adjunct to antibiotics compared to neonates with confirmed gram negative sepsis who had placebo. [typical RR 0.19 (95% CI 0.04, 1.02), typical RD -0.25 (95% CI -0.46, -0.04), NNT 4 (95% CI 2, 25)]. There was no significant heterogeneity in this result for the two trials.

Other planned subgroup analyses:

Both the included trials analysed preterm infants with confirmed, late onset sepsis only. Thus, the planned subgroup analyses based on preterm infants, infants with confirmed sepsis or infants with late onset sepsis are in fact identical to the analyses reported in this review for all infants. The following subgroup analyses were not possible due to lack of data: term infants, early onset sepsis, suspected sepsis or fungal sepsis.

Secondary outcomes

No adverse effects due to pentoxifylline were observed in the two included trials. None of the other secondary outcomes were reported.

II. PENTOXIFYLLINE FOR TREATMENT OF NEC (COMPARISON 02):

No completed randomised or quasi-randomised trials using pentoxifylline for the treatment of stage II or III NEC were identified.

Discussion

This review identified only two eligible RCT's using pentoxifylline for the treatment of neonatal sepsis. Both studies were performed by the same principal investigator in Poland. The two studies included 140 preterm neonates with suspected late onset sepsis. Of these, 33 (23%) of the enrolled subjects were excluded because of unconfirmed sepsis. Thus, the analysis was not based on 'intention to treat'. The trials did not include sample size calculations and studied a very small number of neonates with sepsis. Two ongoing trial using pentoxifylline for NEC that will be completed in 2009 or 2010 were identified (Hammerman 2005, Patole 2006).

In this review, a significant reduction in all cause mortality during hospital stay in neonates with sepsis who were treated with pentoxifylline as an adjunct to antibiotics compared to placebo was noted. In the subgroup analysis, a reduction of borderline statistical significance in all cause mortality during hospital stay in neonates with confirmed gram negative sepsis who had pentoxifylline as an adjunct to antibiotics compared to neonates with confirmed gram negative sepsis who had placebo was noted. This subgroup analysis only included 49 neonates. There was no significant heterogeneity between the results of the included studies (p > 0.10).

Possible haematological adverse effects of pentoxifylline affecting platelet, leukocyte and red cell function were not observed in the included trials. However, pentoxifylline is currently not licensed for use in neonates in many countries and thus may not be widely available.

Despite advances in neonatal care and use of potent antimicrobials, the mortality and morbidity from neonatal sepsis and NEC remain high (Stoll 2005; Stoll 2002; Lin 2006). Adjuvants that modulate inflammation, in conjunction with antibiotics, have the potential to reduce mortality and limit cerebral and pulmonary injury. Intravenous immunoglobulin (IVIG) is one of the agents that modulates inflammation and used to treat or prevent sepsis (Haque 1998). In seven trials (n = 262) of neonates with confirmed sepsis, treatment with IVIG resulted in a statistically significant reduction in mortality; however, Ohlsson and Lacy in their Cochrane review (Ohlsson 2002) concluded that there was insufficient evidence to support the routine administration of IVIG for neonatal sepsis. A large trial (approximately 3500 infants) evaluating IVIG in sepsis (International Neonatal Immunotherapy Study) and funded by the Medical Research Council (UK) which recruited approximately 3500 neonates has recently concluded.

Pentoxifylline has been shown to have numerous potential benefits by modulating inflammation in human and animal models of sepsis and NEC, i.e. suppression of TNF-α, IL-6 and IL-8, and a variety of physiological effects at cellular, vascular and endothelial levels. Thus, it appears to be a promising adjunct in the treatment of neonatal sepsis and NEC. The results from the two studies identified in the this review show a statistically significant reduction in mortality and a trend towards earlier correction of metabolic and haemodynamic derangements in preterm neonates with confirmed late onset sepsis. However, because of the methodological weaknesses of the studies highlighted earlier, the routine use of pentoxifylline cannot be recommended.

Authors' conclusions

Implications for practice

Current evidence from two small RCT's suggests that pentoxifylline as an adjunct to antibiotic therapy in late onset neonatal sepsis significantly reduces mortality during hospital stay without adverse effects. However, these trials have considerable methodological weaknesses and the results of this meta-analysis should be interpreted with caution. No completed trial using pentoxifylline for the treatment of NEC was identified.

Implications for research

Researchers should be encouraged to undertake large well designed RCT's to confirm or refute the role of pentoxifylline in treatment of confirmed or suspected early and late onset neonatal sepsis as well as NEC. The trials should report on clinically important co-morbidities of sepsis (e.g. chronic lung disease, periventricular leukomalacia, duration of assisted ventilation, among others) and long term neurological outcome. Researchers may consider comparing pentoxifylline with other adjunctive modalities to treat neonatal sepsis such as colony stimulating factors or intravenous immunoglobulins.

Acknowledgements

  1. Ms. Nicola Bexon of the Institute of Health Sciences for helping to formulate the literature search strategy
  2. Mr. Edward George of Hirson Library at St Helier Hospital for assisting to formulate the literature search strategy
  3. Miss Natasha Haque for help with typing the manuscript

Contributions of authors

Khalid Haque

Searched the literature and identified trials for inclusion
Assessed methodology
Extracted relevant data from eligible studies
Verified and entered data in RevMan
Contacted prominent authors for more data on published and unpublished trials
Wrote the text of the review
Is the corresponding author
Assisted in updating the review

Pammi Mohan

Assisted in writing the protocol and review
Independently assessed study methodology and extracted data from eligible studies
Assisted in entering and checking data in RevMan
Assisted in contacting authors for more information on published articles and in tracing unpublished ones
Updated the review

Potential conflict of interest

  • None noted.

[top]

Characteristics of studies

Characteristics of Included Studies

Lauterbach 1996

Methods

Single centre, randomised placebo controlled trial. No details of randomisation given.
Blinding of intervention-yes.
Blinding of outcome-yes.
Completeness of follow-up-no.
11/40 patients were excluded from analysis.

Participants

Single centre. Neonatal Unit, Jagiellonan University Hospital, Poland.
Neonates < 36 weeks gestation, with clinically
suspected sepsis after the first week of life.
Criteria for sepsis; at least 2 of the following; feed intolerance, abdominal distention, lethargy, irritability, temperature instability, hyperbilirubinaemia, hepato-splenomegaly.
Exclusions; major congenital malformation, Grade III and IV intra-ventricular haemorrhage and congenital infections.
Mean gestational age in treatment group 31.5 weeks, in placebo group 32.3 weeks.
Mean birth weight in treatment group 1.75 Kg and in placebo group 1.86 Kg.
Period of study 1st March to 30th July 1994.
Only infants with positive blood culture were analysed. Treatment group 16/20 (4 negative culture), placebo group 13/20 (7 negative culture)

Interventions

Pentoxifylline (Trental; Boehring-Hoscht) 5 mg/kg/hour for 6 hours, repeated on 2nd and 3rd day ( n=20).
Placebo: equal volume of normal saline (n=20).

Outcomes

Outcomes reported were;

  1. Mortality
  2. TNF alpha levels before the first infusion and after the third infusion of the drug or placebo.
  3. Adverse reactions to the drug.
Notes
Risk of bias table
Item Judgment Description
Allocation concealment? Unclear

B - Unclear

Lauterbach 1999

Methods

Randomised placebo controlled study.
Concealment of randomisation: yes.
Blinding of intervention: yes.
Blinding of outcome assessment: cannot tell.
Completeness of follow-up: no.
22/100 were excluded.

Participants

Two centre study: Neonatal Unit, Medical College Jagiellonan, University of Cracow, Poland and ITU at Polish Mother and Children Hospital, Lodz, Poland.
Preterm infants < 36 weeks of gestation, after first week of life with suspected sepsis.
Criteria for sepsis: At least three of the following; feeding intolerance, abdominal distention, temperature instability, disordered peripheral circulation (as described by paleness, peripheral cyanosis, mottled skin and capillary refill time > 3 seconds), lethargy, irritability and hepato-splenomegaly.
Positive blood culture required for confirmation of sepsis.
Period: 1st Jan 1995 to 30th July 1996.
Exclusions: Congenital malformations, congenital infections and grade III and IV intra-ventricular haemorrhage.
Total recruitment 100.
10/50 from the treatment group and 12/50 from placebo group excluded as sepsis was not confirmed.

Interventions

Pentoxifylline (Pentilin -KRKA Slovenia) 5 mg/kg/hour for 6 hours for 6 successive days (n=50).
Placebo: Equal volume of Normal saline (n=50).

Outcomes

Outcomes reported were:

  1. Mortality
  2. Incidence and mortality of gram negative sepsis.
  3. Plasma TNF alpha, IL-1 and IL-6 levels before and after the 1st and 3rd dose and before and after the 6th dose.
  4. Adverse effects.
Notes
Risk of bias table
Item Judgment Description
Allocation concealment? Yes

A - Adequate

Characteristics of excluded studies

Lauterbach 1994

Reason for exclusion

Not a randomised or a quasi-randomised trial

Characteristics of studies awaiting classification

  • None noted.

Characteristics of ongoing studies

Hammerman 2005

Study name

Pentoxifylline in the treatment of NEC in premature neonates

Methods
Participants

Premature neonates (< 1750g birthweight) with clinical suspicion of NEC and abdominal X-ray consistent with NEC

Interventions

Pentoxifylline IV at 5mg/kg/hr to run over 6hr for 6 days or an equal volume of 1/2 normal saline as placebo

Outcomes

To be confirmed with the principal investigator.

Starting date

Jan 2005

Contact information

Cathy Hammerman- email - cathy@cc.huji.ac.il

Notes

Finishes in March 2009

Patole 2006

Study name

Safety and efficacy of pentoxifylline as a treatment for preventing the progression of necrotising enterocolitis in preterm neonates– A randomised, placebo controlled pilot trial

Methods
Participants

Premature neonates < 32 weeks gestation with Stage II or III NEC

Interventions

Pentoxifylline at 5mg/kg/hr for 12hrs for 2 days followed by the infusion for 6 hours a day for the next 4 days OR equal volume of placebo in controls

Outcomes

Primary: Efficacy and safety of pentoxifylline in preventing the progression of NEC and or death.
Secondary : Reduction in plasma TNF-alpha levels, extent of boel resection at surgery, duration of hospital stay and TPN support, and duration to full enteral feeds

Starting date

June 2006

Contact information

Sanjay.Patole@health.wa.gov.au

Notes

To finish in 2009 or 2010

[top]

References to studies

Included studies

Lauterbach 1996

Lauterbach R, Zembala M. Pentoxifylline reduces plasma tumour necrosis factor-alpha concentration in premature infants with sepsis. European Journal of Pediatrics 1996;155:404-9.

Lauterbach 1999

Lauterbach R, Pawlik D, Danuta K, Wieslaw K, Ewah K, Marek Z. Effect of immunomodulating agent, pentoxifylline, in the treatment of sepsis in prematurely delivered infants: a placebo controlled, double-blind trial. Critical Care Medicine 1999;27:807-14.

Excluded studies

Lauterbach 1994

Lauterbach R, Pawlik D, Tomaszcyk B, Cholewa B. Pentoxifylline treatment of sepsis of premature infants: preliminary clinical observations. European Journal of Pediatrics 1994;153:672-74.

Studies awaiting classification

  • None noted.

Ongoing studies

Hammerman 2005

Patole 2006

Other references

Additional references

Adams-Chapman 2006

Adams-Chapman I, Stoll BJ. Neonatal infection and long-term neurodevelopmental outcome in the preterm infant. Current Opinion in Infectious Disease 2006;19:290-7.

Bacher 1997

Bacher A, Mayer N, Klimscha W, et al. Effects of pentoxifylline on haemodynamics and oxygenation in septic and non-septic patients. Critical Care Medicine 1997;25:795-800.

Bell 1978

Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, Barton L, Brotherton T. Neonatal necrotising enterocolitis. Therapeutic decisions based upon clinical staging. Annals of Surgert 1978;187:1-7.

Bhutta 1997

Bhutta ZA. Neonatal infections. Current Opinion in Paediatrics 1997;9:133-40.

Bienvenu 1995

Bienvenu J, Doche C, Gutowski MC, Lenoble M, Lepape A, Perdrix JP. Production of pro-inflamatory cytokines and cytokines involved in the TH1/TH2 balance is modulated by pentoxifylline. Journal of Cardiovascular Pharmacology 1995;25:S80-4.

Boldt 1996

Boldt J, Müller M, Heyn S, Welters I, Hempelmann G. Influence of long term continuous intravenous administration of pentoxyfylline on endothelial related coagulation in critically ill patients. Critical Care Medicine 1996;24:940-6.

Caplan 1990

Caplan MS, Hsueh W. Necrotizing enterocolitis: role of platelet activating factor, endotoxin, and tumor necrosis factor. Journal of Pediatrics 1990;117:S47-51.

Caplan 1990a

Caplan MS, Sun XM, Hseuh W, Hageman JR. Role of platelet activating factor and tumor necrosis factor-alpha in neonatal necrotizing enterocolitis. Journal of Pediatrics 1990;116:960-4.

Edelson 1999

Edelson MB, Bagwell CE, Rozycki HJ. Circulating pro- and counterinflammatory cytokine levels and severity in necrotizing enterocolitis. Pediatrics 1999;103:766-71.

Halpern 2006

Halpern MD, Clark JA, Saunders TA, Doelle SM, Hosseini DM, Stagner AM, et al. Reduction of experimental necrotizing enterocolitis with anti-TNF-alpha. American Journal of Physiology. Gastrointestinal and Liver Physiology 2006;290:G757-64.

Haque 1998

Haque KN. In: Forfar and Arneil, editor(s). Textbook of Paediatrics. 5th edition. Edinburgh: Churchill Livingstone, 1998.

Harris 2005

Harris MC, D'Angio CT, Gallagher PR, Kaufman D, Evans J, Kilpatrick L. Cytokine elaboration in critically ill infants with bacterial sepsis, necrotizing entercolitis, or sepsis syndrome: correlation with clinical parameters of inflammation and mortality. Journal of Pediatrics 2005;147:462-8.

Jaattela 1991

Jaattela M. Biologic activities and mechanisms of action of tumor necrosis factor-alpha/cachectin. Laboratory Investigation 1991;64:724-42.

Kaufman 2004

Kaufman D, Fairchild KD. Clinical microbiology of bacterial and fungal sepsis in very-low-birth-weight infants. Clinical Microbiology Reviews 2004;17:638-80.

Krysztopik 1996

Krysztopik RJ, Bentley FR, Wilson MA, et al. Free radical scavenging by lazaroids improves renal blood flow during sepsis. Surgery 1996;20:657-62.

Lawn 2006

Lawn JE, Wilczynska-Ketende K, Cousens SN. Estimating the causes of 4 million neonatal deaths in the year 2000. International Journal of Epidemiology 2006;35:706-18.

Levy 1998

Levy SB. Antimicrobial resistance: Bacteria on the defence. Resistance stems from misguided efforts to try to sterilize our environment. BMJ 1998;317:612-13.

Lin 2006

Lin PW, Stoll BJ. Necrotizing enterocolitis. Lancet 2006;368:1271-83.

Michetti 2003

Michetti C, Coimbra R, Hoyt DB, Loomis W, Junger W, Wolf P. Pentoxifylline reduces acute lung injury in chronic endotoxemia. Journal of Surgical Research 2003;115:92-9.

Ng 2003

Ng PC, Li K, Wong RPO, Chui K, Wong E, Li G, Fok TF. Proinflammatory and anti-inflammatory cytokine responses in preterm infants with systemic infections. Archives of Disease in Childhood Fetal Neonatal Ed 2003;88:F209-13.

Ohlsson 2002

Ohlsson A, Lacy JB. Intravenous immunoglobulin for suspected or subsequently proven infection in neonates. Cochrane Database of Systematic Reviews 2002, Issue 4.

Philip 1994

Philip AGS. The changing face of neonatal infections: experience at a regional medical centre. Pediatric Infectious Disease Journal 1994;13:1098-1102.

Speer 1999

Speer CP. Inflammatory mechanisms in neonatal chronic lung disease. European Journal of Pediatrics 1999;158(suppl):S18-22.

Stoll 2002

Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, Ehrenkranz RA, et al. Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics 2002;110:285-91.

Stoll 2004a

Stoll BJ. Infections of the neonatal infant. In: Behrman RE KR, Jenson HB, editor. Nelson Textbook of Pediatrics. 17th ed edition. Philadelphia: Saunders, 2004:623-40.

Stoll 2004b

Stoll BJ, Hansen NI, Adams-Chapman I, Fanaroff AA, Hintz SR, Vohr B, et al. Neurodevelopmental and growth impairment among extremely low-birth-weight infants with neonatal infection. JAMA 2004;292:2357-65.

Stoll 2005

Stoll BJ, Hansen NI, Higgins RD, Fanaroff AA, Duara S, Goldberg R, et al. Very low birth weight preterm infants with early onset neonatal sepsis: the predominance of gram-negative infections continues in the National Institute of Child Health and Human Development Neonatal Research Network, 2002-2003. Pediatric Infectious Disease Journal 2005;24:635-9.

Travadi 2006

Travadi J, Patole S, Charles A, Dvorak B, Doherty D, Simmer K. Pentoxifylline reduces the incidence and severity of necrotizing enterocolitis in a neonatal rat model. Pediatric Research 2006;60:185-9.

Vilcek 1991

Vilcek J, Lee TH. Tumor necrosis factor. New insights into the molecular mechanisms of its multiple actions. Journal of Biological Chemistry 1991;266:7313-6.

Volpe 1995

Volpe JJ. Neurology of the Newborn. 3rd edition. Philadelphia, London: WB Saunders, 1995.

Walsh 1986

Walsh MC, Kliegman RM. Necrotizing enterocolitis: Treatment based on staging criteria. Pediatric Clinics of North America 1986;33:179-201.

Wang 1996

Wang P, Wood TJ, Ba ZF, Chaudry IH. Pentoxifylline maintains vascular endothelial cell function during hyperdynamic sepsis. Surgery 1996;130:367-73.

Yang 1999

Yang S, Zhou M, Koo DJ, Chaudry IH, Wang P. Pentoxifylline prevents the transition from hyperdynamic to hypodynamic response during sepsis. American Journal of Physiology 1999;277:H1036-44.

Zeni 1996

Zeni F, Pain P, Vindimian M, Gay JP, Gery P, Bertrand M et al. Effects of pentoxifylline on circulating cytokine concentrations and haemodynamics in patients with septic shock: Results from a double blind, randomized, placebo controlled study. Critical Care Medicine 1996;24:207-14.

Other published versions of this review

Haque 2003

Haque K, Mohan P. Pentoxifylline for neonatal sepsis. Cochrane Database of Systematic Reviews 2003, Issue 4.

[top]

Data and analyses

1 Pentoxifylline versus placebo for the treatment of neonatal sepsis

Outcome or Subgroup Studies Participants Statistical Method Effect Estimate
1.1 All cause mortality during hospital stay 2 Risk Ratio (M-H, Fixed, 95% CI) Subtotals only
1.1.1 All infants 2 107 Risk Ratio (M-H, Fixed, 95% CI) 0.14 [0.03, 0.76]
1.1.2 Infants with confirmed gram negative sepsis 2 49 Risk Ratio (M-H, Fixed, 95% CI) 0.19 [0.04, 1.02]

[top]

Sources of support

Internal sources

  • Epsom & St. Helier NHS Trust, UK
  • National Perinatal Epidemiology Unit, Headington, Oxford, UK

External sources

  • No sources of support provided.

This review is published as a Cochrane review in The Cochrane Library, Issue 1, 2008 (see http://www.thecochranelibrary.com External Web Site Policy for information). Cochrane reviews are regularly updated as new evidence emerges and in response to feedback. The Cochrane Library should be consulted for the most recent version of the review.