Treatments for priapism in boys and men with sickle cell disease

Francis I Chinegwundoh; Sherie Smith; Kofi A Anie

doi:10.1002/14651858.CD004198.pub3

Liječenje prijapizma u dječaka s anemijom srpastih stanica

Declaraciones de intereses de los autores

Versión publicada: 19 septiembre 2017 Historial de versiones

https://doi.org/10.1002/14651858.CD004198.pub3

Contraer todo Desplegar todo

Abstract

Background

Sickle cell disease comprises a group of genetic haemoglobin disorders. The predominant symptom associated with sickle cell disease is pain resulting from the occlusion of small blood vessels by abnormally 'sickle‐shaped' red blood cells. There are other complications, including chronic organ damage and prolonged painful erection of the penis, known as priapism. Severity of sickle cell disease is variable, and treatment is usually symptomatic. Priapism affects up to half of all men with sickle cell disease, however, there is no consistency in treatment. We therefore need to know the best way of treating this complication in order to offer an effective interventional approach to all affected individuals.

Objectives

To assess the benefits and risks of different treatments for stuttering (repeated short episodes) and fulminant (lasting for six hours or more) priapism in sickle cell disease.

Search methods

We searched the Cochrane Cystic Fibrosis and Genetic Disorders Group Haemoglobinopathies Trials Register, which comprises references identified from comprehensive electronic database searches and handsearches of relevant journals and abstract books of conference proceedings. We also searched trial registries.

Date of the most recent search of the Group's Haemoglobinopathies Trials Register: 15 September 2017.

Date of most recent search of trial registries and of Embase: 12 December 2016.

Selection criteria

All randomised or quasi‐randomised controlled trials comparing non‐surgical or surgical treatment with placebo or no treatment, or with another intervention for stuttering or fulminant priapism.

Data collection and analysis

The authors independently extracted data and assessed the risk of bias of the trials.

Main results

Three trials with 102 participants were identified and met the criteria for inclusion in this review. These trials compared stilboestrol to placebo, sildenafil to placebo and ephedrine or etilefrine to placebo and ranged in duration from two weeks to six months. All of the trials were conducted in an outpatient setting in Jamaica, Nigeria and the UK. None of the trials measured our first primary outcome, detumescence but all three trials reported on the reduction in frequency of stuttering priapism, our second primary outcome. No significant effect of any of the treatments was seen compared to placebo. Immediate side effects were not found to be significantly different from placebo in the two trials where this information was reported. We considered the quality of evidence to be low to very low as all of the trials were at risk of bias and all had low participant numbers.

Authors' conclusions

There is a lack of evidence for the benefits or risks of the different treatments for both stuttering and fulminant priapism in sickle cell disease. This systematic review has clearly identified the need for well‐designed, adequately‐powered, multicentre randomised controlled trials assessing the effectiveness of specific interventions for priapism in sickle cell disease.

PICO

Population

Intervention

Comparison

Outcome

El uso y la enseñanza del modelo PICO están muy extendidos en el ámbito de la atención sanitaria basada en la evidencia para formular preguntas y estrategias de búsqueda y para caracterizar estudios o metanálisis clínicos. PICO son las siglas en inglés de cuatro posibles componentes de una pregunta de investigación: paciente, población o problema; intervención; comparación; desenlace (outcome).

Para saber más sobre el uso del modelo PICO, puede consultar el Manual Cochrane.

Laički sažetak

disponible en

Liječenje dugotrajne i bolne erekcije penisa (prijapizam) u dječaka s anemijom srpastih stanica

Istraživačko pitanje

Koje su koristi i rizici različitih postupaka liječenja kratkog i ponavljanog (koje se ponavlja u kratkim epizodama) i fulminantnog (u trajanju od šest sati ili više) prijapizma kod anemije srpastih stanica?

Dosadašnje spoznaje

Prijapizam (produljena bolna erekcija penisa) je čest u muškaraca s anemijom srpastih stanica. Trajanje prijapizma razlikuje se kod različitih tipova poremećaja, te stoga postoji i više postupaka liječenja. Pristupi samopomoći mogu biti korisni. Tražili smo randomizirana kontrolirana ispitivanja različitih postupaka liječenja kako bi pronašli najbolju opciju.

Datum pretraživanja

Zadnje pretraživanje literature obavljeno je 15 rujna 2017.

Obilježja uključenih istraživanja

Pronašli smo tri ispitivanja provedena u Jamajci, Nigeriji i Velikoj Britaniji koja su uključivala ukupno 102 osobe.

Ključni rezultati

U kliničkim ispitivanjima, četiri različita lijeka (stilboestrol, sildenafil, efedrin i etilefrine) bila su uspoređivana s placebom. Sva su ispitivanja istraživala da li postupci liječenja smanjuju učestalost napadaja prijapizma koja se događaju. Nije bilo razlike između bilo kojeg postupka liječenja u usporedbi s placebom. Zbog nedostatka dokaza, nismo u mogućnosti zaključiti koji je najbolji postupak liječenja prijapizma kod anemije srpastih stanica. Potrebno je obaviti dodatna istraživanja.

Kvaliteta dokaza

Zaključili smo da je kvaliteta dokaza niska do vrlo niska jer je kod svih ispitivanja uočen rizik od pristranosti i sva su imala mali broj sudionika.

Authors' conclusions

Implications for practice

The quality and quantity of the trials included in this review were not sufficient to allow firm conclusions about treatment for priapism in sickle cell disease to be made at the present time.

Implications for research

Well‐designed, adequately‐powered trials with multicentre collaboration are desperately required to provide evidence for the most effective treatment required to alleviate suffering in people with sickle cell disease. Large numbers of participants would be required in a randomised clinical trial to provide the evidence base for drug therapy of priapism and to demonstrate that one regimen is more efficacious than another. In acute or fulminant priapism, it would be unethical to have a placebo‐controlled (randomised) trial. However, trials could be conducted comparing one agent against another, e.g. different alpha agonists.

Summary of findings

Open in table viewer

Summary of findings for the main comparison. Silboesterol versus placebo

Silboesterol compared with placebo for stuttering priapism
Patient or population: men and boys with SCD and stuttering priapism Settings: outpatients Intervention: stilboestrol 5 mg Comparison: placebo
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No. of Participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Stilboestrol 5 mg
Detumescence Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Reduction in stuttering priapism Follow‐up: 2 weeks	250 per 1000	730 per 1000 (183 to 1000)	RR 2.92 (95% CI 0.73 to 11.70)	7 (1)	⊕⊝⊝⊝ very low^1,2
Immediate side effects of treatment Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Effect on later sexual function Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Other untoward side effects of treatment Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Efficacy of a prevention strategy Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
The basis for the assumed risk* (e.g. the median control group risk across trials) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; N/A: not applicable; RR: risk ratio.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.
1. Downgraded once due to unclear methods of randomisation and allocation. 2. Downgraded twice for imprecision: confidence intervals around the relative effect are very wide as the trial has a small number of participants and a low event rate.

Open in table viewer

Summary of findings 2. Sildenafil versus placebo

Sildenafil 50 mg compared with placebo for stuttering priapism
Patient or population: men and boys with SCD and stuttering priapism Settings: outpatients Intervention: sildenafil 50 mg daily Comparison: placebo
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No. of Participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Sildenafil 50 mg daily
Detumescence Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Reduction in frequency of priapism Follow‐up: 8 weeks	286 per 1000	166 per 1000 (20 to 1000)	RR 0.58 (95% CI 0.07 to 4.95)	13 (1)	⊕⊕⊝⊝ low^1,2	There was also no significant difference between treatments for the reduction in episodes of priapism by score tier: RR 1.17 (95% CI 0.36 to 3.76).
Immediate side effects of treatment Follow‐up: 16 weeks	See comment	See comment	N/A	13 (1)	⊕⊕⊝⊝ low^1,3	Side effects of treatment were reported for the whole treatment phase including the open label phase. No significant differences were found between sildenafil and placebo.
Effect on later sexual function Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Other untoward side effects of treatment Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Efficacy of a prevention strategy Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
The basis for the assumed risk* (e.g. the median control group risk across trials) is provided in footnotes. The corresponding risk (and its 95% CI is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; N/A: not applicable; RR: risk ratio.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.
1. Downgraded once due to unclear methods of randomisation and allocation. 2. Downgraded once due to imprecision. Confidence intervals around the relative effect are very wide as the trial has a low number of participants. 3. Downgraded once due to indirectness as we are only interested in the treatment phase but they have included the open label phase for reporting of adverse effects.

Open in table viewer

Summary of findings 3. Etilefrine versus placebo

Etilefrine 50 mg compared with placebo for stuttering priapism
Patient or population: men and boys with SCD and stuttering priapism Settings: outpatient Intervention: etilefrine 50mg once daily Comparison: placebo
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No. of Participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Etilefrine 50 mg
Detumescence Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Reduction in frequency of priapism Follow‐up: 6 months	See comment	See comment	N/A	23 (1)	⊕⊝⊝⊝ very low^1,2	No significant difference was found between etilefrine and placebo but there were concerns over the reliability of the published results.
Immediate side effects of treatment (palpitations) Follow‐up: 6 months	333 per 1000	546 per 1000 (206 to 1000)	RR 1.64 (95% CI 0.62 to 4.30)	23 (1)	⊕⊝⊝⊝ very low^1,2	No significant difference was seen between etilefrine and placebo for any of the other immediate side effects measured (lack of sleep, hand shaking, anxiety, dry mouth).
Immediate side effects of treatment (tachycardia) Follow‐up: 6 months	250 per 1000	545 per 1000 (178 to 1000)	RR 2.18 (95% CI 0.71 to 6.68)	23 (1)	⊕⊝⊝⊝ very low^1,2	No significant difference was seen between etilefrine and placebo for any of the other immediate side effects measured (lack of sleep, hand shaking, anxiety, dry mouth).
Effect on later sexual function Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Other untoward side effects of treatment Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Efficacy of a prevention strategy Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
The basis for the assumed risk* (e.g. the median control group risk across trials) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; N/A: not applicable; RR: risk ratio.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.
1. Downgraded twice due to high risk of bias from incomplete outcome data and unclear risk of bias from randomisation and allocation concealment. 2. Downgraded once due to imprecision:confidence intervals around the relative effect are very wide as the trial has a low number of participants.

Open in table viewer

Summary of findings 4. Ephedrine versus placebo

Ephedrine 15 mg compared with placebo for stuttering priapism
Patient or population: men and boys with SCD and stuttering priapism Settings: outpatients Intervention: ephedrine 15 mg/30 mg once daily Comparison: placebo
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No. of Participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Placebo	Ephedrine
Detumescence Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Reduction in frequency of priapism Follow‐up: 6 months	See comment	See comment	N/A	24 (1)	⊕⊝⊝⊝ very low^1,2	No significant difference was found between ephedrine 15mg and placebo or ephedrine 30mg and placebo but there were concerns over the reliability of the published results.
Immediate side effects of treatment with 30 mg ephedrine (palpitations) Follow‐up: 6 months	333 per 1000	333 per 1000 (107 to 1000)	RR 1.00 (95% CI 0.32 to 3.10)	24 (1)	⊕⊝⊝⊝ very low^1,2	No significant difference was seen between ephedrine 15mg or 30mg and placebo for any of the other immediate side effects measured (lack of sleep, hand shaking, anxiety, dry mouth).
Immediate side effects of treatment with 30 mg ephedrine (tachycardia) Follow‐up: 6 months	250 per 1000	168 per 1000 (33 to 825)	RR 0.67 (95% CI 0.13 to 3.30)	24 (1)	⊕⊝⊝⊝ very low^1,2	No significant difference was seen between ephedrine 15mg or 30mg and placebo for any of the other immediate side effects measured (lack of sleep, hand shaking, anxiety, dry mouth).
Effect on later sexual function Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Other untoward side effects of treatment Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
Efficacy of a prevention strategy Follow‐up: N/A	See comment	See comment	N/A	N/A	N/A	This outcome was not measured.
The basis for the assumed risk* (e.g. the median control group risk across trials) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; N/A: not applicable; RR: risk ratio.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.
1. Downgraded twice due to high risk of bias from incomplete outcome data and unclear risk of bias from randomisation and allocation concealment. 2. Downgraded once due to imprecision: confidence intervals around the relative effect are very wide as the trial has a low number of participants.

Background

Description of the condition

Sickle cell disease (SCD) is a genetic haemoglobin disorder, caused by inheritance from both parents of an altered beta‐globin chain gene of which one at least is beta S. SCD affects people originating from Sub‐Saharan Africa, the African diaspora, Arab countries, the Mediterranean and South America (Davies 1997). Approximately 75,000 people in the USA (mostly African‐Americans) (NHLBI 2002), and about one in 60 West Africans (WHO 1994) are affected by SCD. There are on average 160 children born with SCD in England each year (Hickman 1999).

The pathophysiology of SCD is caused by the abnormal haemoglobin within red blood cells causing deformation or sickling of the cells when they give up oxygen. These abnormal cells block blood vessels causing vaso‐occlusion, resulting in symptoms and organ damage.

Priapism is the persistence of erection that does not result from sexual desire and fails to subside despite orgasm. There are two kinds of priapism: low‐flow ischaemic or veno‐occlusive priapism, which is the form seen in SCD and high‐flow priapism or non‐ischaemic, which is associated with external trauma causing damage to the cavernosal artery. Priapism is usually accompanied by pain and tenderness. It can occur in all age groups, with a reported peak incidence in SCD between the ages of 5 to 10 years and 20 to 50 years (Lue 1998). However, work from Jamaica suggests that the onset of priapism can be anytime between the ages of 5 and 45 years, with the peak incidence among young adults (aged 20 years to 25 years) (Emond 1980). Typically, priapism only affects the corpora cavernosa (very rarely the corpus spongiosum may be affected) (Lue 1998), thus surgical treatments to relieve priapism are designed to remove blood from the corpora cavernosa. Low‐flow priapism may result in impotence or erectile dysfunction secondary to corporeal fibrosis.

There is no clear demarcation between a prolonged erection and priapism. However, many would call an erection lasting over six hours fulminant or prolonged priapism. This is because in the low‐flow type (as occurs in the priapism of SCD) penile ischaemia and acidosis occur after six hours (Broderick 1994). Others, however, call an erection in males with SCD a priapic episode when it is in excess of one hour, or when it is painful.

Priapism is a common occurrence in SCD and may occur as acute episodes or the so‐called stuttering priapism, defined as recurrent episodes of prolonged erections. These can last from a few minutes to three hours. Stuttering priapism by repetitive vaso‐occlusion can lead to erectile dysfunction, and may develop into full‐blown priapism. Treatment is aimed at preventing such an occurrence. In one study of men with homozygous SCD (SS) in a Jamaican clinic, stuttering nocturnal attacks of priapism lasting two to six hours affected 42 per cent of participants (Serjeant 1985). A more recent questionnaire study from the USA reported an actuarial probability of experiencing priapism by the age of 20 years as 89% (+/‐ 9%) (Mantadakis 1999).

Red cell sickling and later sludging of blood occur in the corpora cavernosa, perhaps as a result of abnormal endothelial adherence, the relatively acidic state of the corpora during erection, mild acidosis accompanying hypoventilation during sleep, or mild trauma with masturbation and intercourse. When the venous channels are maximally compressed during nocturnal penile tumescence, the sludged red cells can block the microscopic subtunical venules and trigger diffuse veno‐occlusion (Lue 1998). More recent evidence has shown that priapism results from deficient erection control mechanisms at a molecular level (Bivalacqua 2006).

Description of the intervention

Interventions for priapism fall into two categories of preventing or reducing the frequency of priapic attacks, and treating fulminant priapic episodes. Several treatments have been described including: analgesia; hydration; blood alkalinisation; red cell transfusion; partial exchange transfusion to lower haemoglobin S; pharmaceutical agents and surgical procedures such as washouts. Surgical shunt procedures and even penile implantation have been described for intractable cases (Upadhyay 1998). Current treatments focus on resolving acute attacks and are often administered too late during an episode. With improved understanding of the true pathophysiology of priapism in SCD, newer treatments aim to target disease specific molecular abnormalities. Recent research suggests that there is abnormal signalling of the endothelium‐derived nitric oxide and phosphodiesterase type 5 (PDE5) signal transduction pathway in the penis (Bivalacqua 2012).

How the intervention might work

With a better understanding of the underlying pathophysiology of priapism in men with SCD, interventions can be better targeted at the underlying cause. Interventions to treat an acute attack aim to abort the attack and restore normal vascular homeostasis. It is thought that the nitric oxide imbalance leads to PDE5 dysregulation which then has a physiological impact on cyclic guanosine monophosphate (cGMP) and cyclic adenosine monophosphatecAMP activity. Treatments that interrupt abnormal mechanisms may be effective in reducing the frequency of attacks.

Why it is important to do this review

Priapism is associated with with long‐term erectile dysfunction and psychological consequences in often young men with SCD. It is important to review the evidence for interventions to treat acute episodes of priapism but also to prevent or reduce episodes in the longer term. Treatment of acute and stuttering priapism often depends on the type of clinician or institution to whom the male with SCD presents. With an increasing understanding of the underlying pathophysiology of the condition, new treatments such as etilefrine, an α adrenergic stimulator, and sildenafil, are showing encouraging results for preventing priapism (Bachir 1996; Okpala 2002; Virag 1996). This review seeks to determine safe, effective interventions for preventing or treating (or both) priapism.

Objectives

To assess the benefits and risks of the different treatments for both stuttering and fulminant priapism in SCD.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) and quasi‐randomised trials, whether published or unpublished.

Types of participants

Males with priapism, either fulminant or stuttering with any type of SCD including: homozygous SCD, i.e. sickle cell anaemia (SS); haemoglobin SC disease (SC); sickle beta thalassaemia (Sβº and Sβ⁺), irrespective of age, race, ethnic origin and setting.

Types of interventions

Any treatment for priapism in SCD, whether fulminant or stuttering, non‐surgical or surgical, compared with placebo or no treatment, or against another intervention.

Types of outcome measures

Primary outcomes

Detumescence or resolution of priapism and return of penis to non‐erect state (immediate or after one, two or four hours)
Reduction in frequency of stuttering priapism

Secondary outcomes

Return of priapism
Immediate side effects, e.g. pain, bleeding
Later sexual function
Other untoward systemic side effects, e.g. gynaecomastia, liver problems
Efficacy of a prevention strategy for stuttering priapism (and thus preventing fulminant priapism)

Search methods for identification of studies

There were no restrictions regarding language or publication status.

Electronic searches

Relevant trials were identified from the Group's Haemoglobinopathies Trials Register using the terms: (sickle cell OR (haemoglobinopathies AND general)) AND (priap* OR erection).

The Haemoglobinopathies Trials Register is compiled from electronic searches of the Cochrane Central Register of Controlled Trials (CENTRAL) (updated each new issue of the Cochrane Library) and weekly searches of MEDLINE. Unpublished work is identified by searching the abstract books of five major conferences: the European Haematology Association conference; the American Society of Hematology conference; the British Society for Haematology Annual Scientific Meeting; the Caribbean Public Health Agency Annual Scientific Meeting (formerly the Caribbean Health Research Council Meeting); and the National Sickle Cell Disease Program Annual Meeting. For full details of all searching activities for the register, please see the relevant section of the Cochrane Cystic Fibrosis and Genetic Disorders Group's website.

Date of the most recent search of the Group's Haemoglobinopathies Trials Register: 15 September 2017.

We also searched Embase separately (1974 to 2003) (Appendix 1) and the Internet (December 2003) (Appendix 2) for the last update of this review. For this update Embase is included in the CENTRAL search but there may be a delay before references appear in CENTRAL. We therefore searched Embase for the six months prior to the main search being carried out (16 August 2016) to current (12 December 2016) (Appendix 3).

Date of most current Embase search: 12 December 2016.

We searched the clinical trial registries ClinicalTrials.gov (http://www.clinicaltrials.gov) and the World Health Organisation International Clinical Trials Registry Platform (http://www.who.int/ictrp/search/en/) for any ongoing and or unpublished trials using the terms " sickle cell" and "priapi*" (Appendix 4).

Date of most recent search 12 December 2016.

Searching other resources

Reference lists

We searched reference list of all included trials to identify any further relevant trials.

Date of the latest search: 12 December 2016.

Correspondence

Attempts were also made to identify any unpublished trials through contact with experts in the field and personal communication with known authors.

Data collection and analysis

Selection of studies

Each author independently identified potentially relevant trials from the search results. No disagreement occurred, so no other expert in the field was invited to make an independent assessment based on the selection criteria.

Data extraction and management

Two authors independently extracted data using standard data acquisition forms. Where there was disagreement or uncertainty, a third author arbitrated.

Outcome data were to be grouped into those measured at 1, 3, 6 and 12 months and annually thereafter. This was to gauge the long‐term benefit or risk from an intervention. Trials reporting on time periods not listed above, were considered as well and described within the review.

Assessment of risk of bias in included studies

Two review authors independently assessed trials following the domain‐based evaluation as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). The assessments were compared and any inconsistencies between the review authors were discussed and resolved.

Assessment was made of the following domains, each being assessed as having either a low, unclear or high risk of bias:

generation of the allocation sequence
concealment of allocation
blinding (of participants, personnel and outcome assessors)
incomplete outcome data
1. record of participants lost to follow‐up or subsequently excluded from the trial
2. whether an intention‐to‐treat analyses has been performed
selective reporting
other bias

Measures of treatment effect

For binary outcome measures we calculated a pooled estimate of the treatment effect for each outcome across trials using the risk ratio (RR) with 95% confidence intervals (CIs), where appropriate.

For continuous outcomes, we planned to record either mean change from baseline for each group or mean post‐treatment or post‐intervention values and standard deviation for each group. We also intended to compute a pooled estimate of treatment effect by calculating the mean difference (MD) with 95% CIs; however; we were unable to do this because of insufficient data.

Unit of analysis issues

Elbourne discusses methods for meta‐analysing cross‐over trials (Elbourne 2002). The methods discussed rely on the data that is reported within the primary paper. In this review, we have adopted the method that uses the data from the first period only for the included cross‐over trial, as this is currently the only data available from the trial (Serjeant 1985). However, for future updates, we plan to use the most appropriate method that is available in RevMan to analyse cross‐over data (Review Manager 2014).

Dealing with missing data

We contacted authors, where further clarification was needed.

In order to allow an intention‐to‐treat analysis, we sought data on the number of participants with each outcome event, by allocated treated group, irrespective of compliance and whether or not the participant was later thought to be ineligible or otherwise excluded from treatment or follow‐up.

Assessment of heterogeneity

We could not test for heterogeneity between trial results using a standard Chi² test as intended as the included trials reported different outcomes and could not be combined. However, for future updates, when more trials are included we plan to test for heterogeneity using the Chi² test.

Assessment of reporting biases

We planned to generate a funnel plot to attempt to identify any publication bias in the included trials (Higgins 2011), however, there were less than 10 included trials and we were therefore unable to conduct this analysis. We attempted to identify any selective reporting in the included publications, by comparing the trial protocols or trial registry, where available, with the final papers and by examination of the trial publications. We considered reporting of both positive and negative effects of the intervention. Where trial protocols weren't available, we compared the outcomes reported in the results section against the methods section of the paper. We extracted information on the sponsors, sources of funding and competing interests of the authors to rule out external bias.

Data synthesis

We generated forest plots using the Review Manager software (Review Manager 2014) from the trial data but were unable to combine the results into a meta‐analysis due to the included trials reporting different interventions and outcomes. We carried out separate analyses for each of the interventions.

We have used the fixed‐effect approach. If further trials are included, in the absence of homogeneity of treatment effects, we will use a random‐effect approach, otherwise, we will use the fixed‐effect model.

Subgroup analysis and investigation of heterogeneity

We were unable to carry out subgroup analyses as there were insufficient trials with comparable outcomes. The management of stuttering priapism in people with SCD is about prevention, and therefore differs from the management of fulminant priapism. Thus, future analyses will not only be by group but also by acute management versus prevention.

Sensitivity analysis

We planned to carry out sensitivity analyses to look at the effect of the risk of bias findings and to look at the effect of adding in and taking out trials where there was high risk of bias. This was not possible due to there not being any trials included in a meta‐analysis.

Summary of findings tables and quality of the evidence (GRADE)

In a post hoc change in line with current Cochrane guidance, at the 2017 update we added a summary of findings table for each comparison presented in the review (summary of findings Table for the main comparison; summary of findings Table 2; summary of findings Table 3; summary of findings Table 4). We selected the following outcomes to report (chosen based on relevance to clinicians and consumers).

Detumescence
Reduction in stuttering priapism
Immediate side effects of treatment (up to two specific effects)
Effect on later sexual function
Other untoward side effects of treatment
Efficacy of a prevention strategy

We determined the quality of the evidence using the GRADE approach; and downgraded evidence in the presence of a high risk of bias in at least one trial, indirectness of the evidence, unexplained heterogeneity or inconsistency, imprecision of results, high probability of publication bias. We downgraded evidence by one level if they considered the limitation to be serious and by two levels if very serious.

Results

Description of studies

Results of the search

The previous version of the review included one trial (two references) (Serjeant 1985). The 2017 searches identified eight new references, one of which was not eligible for any section of the review. The remaining seven new references were to two new eligible trials (Burnett 2014; Olujohungbe 2011). One of these references was to an ongoing trial that appeared in the previous version of this review but the full paper has now been published and we can include it (Olujohungbe 2011). The results of the search are shown in a PRISMA flow diagram (Figure 1).

Figure 1

Study flow diagram.

Included studies

Nine references to three trials (with a total of 102 participants) on the treatment of stuttering priapism, were identified as eligible for inclusion in the review (Burnett 2014; Olujohungbe 2011; Serjeant 1985) (Characteristics of included studies). The two more recent trials are both double‐blind, placebo‐controlled parallel trials with treatment periods ranging from eight weeks (Burnett 2014) to 26 weeks (Olujohungbe 2011). The third trial is a double‐blind, placebo controlled cross‐over trial, in which participants who did not respond to the first arm of the trial were entered into the second arm of the trial with no washout period described (Serjeant 1985). There were no randomised trials of treatments for fulminant priapism. The literature reveals several case reports and case series of well‐established therapies and experimental therapies for stuttering and fulminant priapism.

Participants and setting

The 102 participants across the three trials were all outpatients with confirmed SCD and stuttering priapism from clinics in the USA, Nigeria, the UK and Jamaica. Olujohungbe recruited from clinics in the middle east for a short period due to poor recruitment in the UK (Olujohungbe 2011). The inclusion criteria for the Burnett and Serjeant trials stated that participants had to have at least two stuttering priapism episodes per week (Burnett 2014; Serjeant 1985). The Olujohungbe trial did not specify the number of episodes of stuttering priapism per week, but did exclude participants who had an attack which lasted longer than four hours or required hospitalisation (Olujohungbe 2011).

The mean (SD) age of participants ranged from 21.7 (5.3) years (Burnett 2014 Sildenafil group) to 24.3 (3.1) years (Serjeant 1985). Participant numbers were low in two of the trials with Burnett including 13 participants and Serjeant including only 11 (Burnett 2014; Serjeant 1985). The Olujohungbe trial required 320 participants to give 80% power to detect differences of 1.5 attacks per month between groups and did include larger numbers of participants but they only enrolled 131 of the required 320 and only 78 were eligible to be randomised to the intervention phase. Only 46 of these participants were included in the analysis and only 25 completed the treatment period (Olujohungbe 2011).

Type of intervention

All three trials report on different treatments from each other but they all compare to a placebo. The Serjeant trial looks at the preventive effects of 5 mg of stilboestrol on stuttering priapism (Serjeant 1985). Burnett looks at the effect of sildenafil 50 mg, a phosphodiesterase type 5 inhibitor, given once daily over eight weeks on preventing attacks of priapism (Burnett 2014). Olujohungbe looks at the effect of two oral alpha‐adrenergic agonists against placebo, etilefrine 50 mg, and two different doses of ephedrine (15 mg and 30 mg) (Olujohungbe 2011). In this trial the participants completed diaries in an observation phase for three to thirteen weeks before taking part in the randomised intervention phase. During the intervention phase participants were asked to take either 15 mg ephedrine, 30 mg ephedrine, 50 mg etilefrine or a placebo once daily, at bedtime for six months.

Type of outcomes

None of the included trials looked at our main primary outcome of detumescence or resolution of a priapism attack but all three trials measured the effect of treatment on the frequency of attacks. Burnett used a scoring system to record the frequency of attacks where 0 = no episodes, 1 = one to two episodes, 2 = three to four episodes, 3 = five to eight episodes and 4 = nine to sixteen episodes and also reported on the duration of episodes (Burnett 2014).The trial of stilboestrol also reported on the return of priapism after a period of no attacks (Serjeant 1985). The two more recent trials reported on immediate side effects of treatment but the specific side effects recorded were different in the two trials due to the actions of the medications being different. Burnett reported on headache, flushing, nasal congestion, abnormal vision, dyspepsia, myalgia whilst Olujohungbe reported on palpitations, tachycardia, lack of sleep, hand shaking, anxiety and dry mouth (Burnett 2014; Olujohungbe 2011). The ephedrine and etilefrine trial was the only trial to look at longer term effects by recording the effect on systolic and diastolic blood pressure but this was as a safety mechanism. Where participants were found to have an increase in systolic blood pressure of 20 mmHg and an increase in diastolic pressure of 10 mmHg the trial medication was discontinued (Olujohungbe 2011).

Excluded studies

There were no excluded studies.

Risk of bias in included studies

The risk of bias was assessed by both authors for all trials included in the review and the details are given in the table Characteristics of included studies. This has been summarised in Figure 2.

Figure 2

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

Allocation

The method of randomisation was unclear and the concealment of allocation was not described in any of the three trials, therefore, we assessed each of these domains as having an unclear risk of bias for all three trials.

Blinding

All three trials were described as double‐blind (physician and participant) and we therefore assessed this domain as low risk. In the Olujohungbe trial, blinding of the trial drugs was done by an external clinical trial supplies company. In the Burnett trial, adverse events were reported for the whole of the treatment period (confirmed by the author team) including the open label phase and we therefore have not entered the data into our analysis as this would bias the results (Burnett 2014).

Incomplete outcome data

The Burnett trial included 13 participants and analysed the data both on an intention‐to‐treat basis and on a per protocol basis. All 13 participants' data were included in the analysis and the trial was deemed to be at low risk of bias for this domain (Burnett 2014).

Although the trial of ephedrine and etilefrine had higher numbers of participants, there was a high rate of attrition at each phase of the trial. The authors stated that 320 participants were needed to give 80% power to detect differences of 1.5 attacks per month between groups. The number enrolled was much less than this (131) and only a proportion completed the observation phase thus enabling them to be randomised to the treatment phase. Of the 78 participants that were randomised to the intervention phase, data was only available for 46 of them. The remainder were lost to follow up. Just under half of the participants analysed completed the full 26 weeks of treatment. We deemed this trial to be at a high risk of bias due to attrition (Olujohungbe 2011).

In the cross‐over trial one participant defaulted, and in two participants the attacks of stuttering priapism ceased spontaneously during the baseline observation period. There was a fourth participant who initially did not take the placebo because a painful crisis aborted the stuttering priapism prior to taking the assigned tablet. In this participant (identified as patient nine) attacks of priapism recurred two weeks later and data on a second baseline period were collected. The paper did not discuss an intention‐to‐treat analysis and we therefore assessed this paper as being at an unclear risk of bias for this domain (Serjeant 1985).

Selective reporting

We deemed two of the trials to be at low risk of bias for selective reporting (Olujohungbe 2011; Serjeant 1985) although we were unable to find a published protocol for any of the trials. The Burnett trial stated that there were no significant differences between groups for decreased median weekly change in priapism episode score but no data were shown (Burnett 2014). The same paper reported adverse effects in the published paper but did not include this as an outcome in either the methods section or in the trial registration document. We deemed this trial to be at an unclear risk of bias for this domain.

Other potential sources of bias

We did not find any other potential sources of bias.

Effects of interventions

See: Summary of findings for the main comparison Silboesterol versus placebo; Summary of findings 2 Sildenafil versus placebo; Summary of findings 3 Etilefrine versus placebo; Summary of findings 4 Ephedrine versus placebo

Silboesterol versus placebo

One trial with 11 participants reported on the effect of 5 mg stilboestrol versus placebo (Serjeant 1985).

Primary outcomes

1. Detumescence (immediate or after one, two or four hours)

This outcome was not assessed in this trial.

2. Reduction in frequency of stuttering priapism

The trial of stilboestrol reported that in four out of the five participants randomised initially to placebo, the attacks of stuttering priapism continued unchanged. In the remaining one participant (participant seven) the attacks ceased immediately on placebo, but he developed a painful crisis six days later. Attacks of priapism recurred one month later and data for a second baseline period were collected before being started on stilboestrol. In all these five participants, switching over to stilboestrol led to the cessation of attacks, immediately in four and gradually in one.

Analysis of the first treatment period (in which no tablet was taken by four participants) showed no significant treatment effect, RR 2.92 (95% CI 0.73 to 11.70) (P = 0.13) (very low quality evidence) (Analysis 1.1). However, the fact that four participants did not take a tablet strongly biases the result (Serjeant 1985).

Secondary outcomes

1. Return of priapism

Return of priapism was addressed in the Serjeant trial but not in a randomised manner (Serjeant 1985).

2. Immediate side effects, e.g. pain, bleeding

This outcome was not assessed (Serjeant 1985).

3. Later sexual function

This outcome was not assessed (Serjeant 1985).

4. Other untoward systemic side effects, e.g. gynaecomastia, liver problems

This outcome was not assessed (Serjeant 1985).

5. Efficacy of a prevention strategy for stuttering priapism (and thus preventing fulminant priapism)

This outcome was not assessed (Serjeant 1985).

Sildenafil versus placebo

One trial with 13 participants reported the effects of sildenafil 50 mg versus placebo (Burnett 2014).