Scolaris Content Display Scolaris Content Display

Intervenciones paliativas para el control de la hemorragia vaginal en el cáncer de cuello de útero avanzado

Contraer todo Desplegar todo

Antecedentes

Esta es una versión actualizada de la revisión Cochrane original publicada en el número 5, 2015.

El cáncer de cuello uterino es el cuarto cáncer más frecuente entre las mujeres en todo el mundo; se calculan 569 847 nuevos diagnósticos y 311 365 muertes por año. Sin embargo, la incidencia y el estadio en el momento del diagnóstico varían enormemente entre las zonas geográficas, y dependen en gran parte de la disponibilidad de un programa de cribado consistente en la población. Por ejemplo, en Nigeria es frecuente la enfermedad en estadio avanzado en el momento de la consulta (86% a 89,3% de los casos nuevos), mientras que en el Reino Unido, solamente el 21,9% de las pacientes se presentan con enfermedad en estadio II+ de la International Federation of Gynaecology and Obstetrics (FIGO). Las pacientes con cáncer de cuello de útero avanzado a menudo necesitan paliación para síntomas angustiantes como la hemorragia vaginal. La hemorragia vaginal puede ser potencialmente mortal en la enfermedad avanzada y su incidencia varía del 0,7% al 100%. La hemorragia es la causa inmediata de muerte en el 6% de las pacientes con cáncer de cuello de útero y su tratamiento a menudo representa un reto.

Por lo tanto, la hemorragia vaginal todavía es una consecuencia frecuente del cáncer de cuello de útero avanzado. Actualmente no hay revisiones sistemáticas que analicen las intervenciones paliativas para el control de la hemorragia vaginal causada por el cáncer de cuello de útero avanzado. Se necesita una evaluación sistemática de las intervenciones paliativas disponibles para informar la toma de decisiones.

Objetivos

Evaluar la eficacia y la seguridad del ácido tranexámico, el taponamiento vaginal (con o sin compresas embebidas en formalina), la radiología intervencionista u otras intervenciones en comparación con radioterapia para el tratamiento paliativo de la hemorragia vaginal en pacientes con cáncer de cuello de útero avanzado.

Métodos de búsqueda

La búsqueda de la revisión original se realizó el 23 de marzo 2015, y las búsquedas posteriores de esta actualización se realizaron el 21 de marzo 2018. Se hicieron búsquedas en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL; 2018, Número 3) en la Cochrane Library; MEDLINE vía Ovid hasta la semana 2 de marzo 2018; y Embase vía Ovid hasta la semana 12 de marzo 2018. También se realizaron búsquedas en registros de ensayos clínicos, resúmenes de reuniones científicas, listas de referencias de artículos de revisión y se estableció contacto con expertos en el campo. Se realizaron búsquedas manuales en las listas de referencias de los estudios relevantes.

Criterios de selección

Para la revisión se realizaron búsquedas de estudios comparativos aleatorizados y no aleatorizados que evaluaron la eficacia y la seguridad del ácido tranexámico, el taponamiento vaginal (con o sin compresas embebidas en formalina), la radiología intervencionista u otras intervenciones, en comparación con técnicas de radioterapia, para el tratamiento paliativo de la hemorragia vaginal en pacientes con cáncer de cuello de útero avanzado (con o sin metástasis), independientemente del estado de publicación, el año de publicación o el idioma.

Obtención y análisis de los datos

Dos autores de la revisión evaluaron de forma independiente si los estudios relevantes cumplían los criterios de inclusión. No se encontraron estudios para inclusión y, por lo tanto, no se analizaron datos.

Resultados principales

La estrategia de búsqueda identificó 1522 referencias únicas, de las cuales se excluyeron 1330 sobre la base del título y el resumen. Se recuperaron los 22 artículos restantes completos, pero ninguno cumplió los criterios de inclusión. Solo se identificaron datos observacionales de estudios de brazo único en pacientes tratadas con compresas embebidas en formalina, radiología intervencionista o técnicas de radioterapia para el control paliativo de la hemorragia vaginal en pacientes con cáncer de cuello de útero.

Conclusiones de los autores

No se encontraron estudios nuevos desde la última versión de esta revisión. No existe evidencia de ensayos controlados para apoyar o refutar el uso de cualquiera de las intervenciones propuestas en comparación con radioterapia. Por lo tanto, la elección de la intervención se debe basar en los recursos locales. Las técnicas de radioterapia para el control de la hemorragia vaginal no son de fácil acceso en ámbitos de escasos recursos, en los que los casos avanzados de cáncer de cuello de útero son predominantes. Por lo tanto, esta revisión sistemática identificó la necesidad de un ensayo controlado aleatorizado que evalúe los efectos beneficiosos y los riesgos de los tratamientos paliativos para la hemorragia vaginal en pacientes con cáncer de cuello de útero avanzado.

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.

¿El taponamiento vaginal, el ácido tranexámico, la radiología intervencionista u otras intervenciones controlan la hemorragia vaginal en pacientes con cáncer de cuello de útero avanzado?

Antecedentes: el cáncer de cuello de útero es el segundo cáncer más frecuente entre las mujeres a nivel mundial, y representa alrededor de 569 847 casos nuevos detectados y 311 365 muertes cada año. Habitualmente las pacientes se presentan a la consulta con enfermedad avanzada en los países en desarrollo, en los que el acceso a los programas de detección del cáncer de cuello de útero es limitado. El cáncer de cuello de útero avanzado puede no ser curable y las pacientes a menudo necesitan tratamiento para el control de síntomas angustiantes (paliación), como la hemorragia vaginal. La hemorragia puede ser tan grave como para ser potencialmente mortal en las pacientes con cáncer de cuello de útero avanzado. El tratamiento de la hemorragia vaginal a menudo representa un reto, especialmente en los países en desarrollo en los que el acceso a la radioterapia es limitado. Las opciones para el tratamiento paliativo de la hemorragia vaginal grave incluyen tratamiento con radiología intervencionista (mediante radiografías para guiar la inserción de "tapones" en los vasos sanguíneos que irrigan el tumor) o taponamiento vaginal (donde se inserta una gasa compactada en la vagina para absorber la sangre y aplicar presión directamente al cuello del útero), aunque estas opciones suelen ser sólo parcialmente efectivas y pueden causar efectos perjudiciales. Las compresas vaginales pueden estar embebidas en formalina, que es un preservante químico. Otras opciones para tratar la hemorragia vaginal grave incluyen ácido tranexámico (un fármaco que reduce la hemorragia y que se puede administrar por vía oral o por inyección) y radioterapia (tratamiento con rayos X de alta energía).

Pregunta de la revisión:el objetivo de esta revisión fue comparar el ácido tranexámico, el taponamiento vaginal (con o sin compresas embebidas en formalina), la radiología intervencionista u otras intervenciones versus el tratamiento con radioterapia para el control de la hemorragia vaginal en el cáncer de cuello de útero.

Principales resultados: las búsquedas se actualizaron hasta marzo 2018. No se encontraron ensayos controlados aleatorizados (estudios clínicos en los que las personas son asignadas al azar a uno de dos o más grupos de tratamiento) para inclusión, por lo que no existe evidencia de que el ácido tranexámico, el taponamiento vaginal (con o sin compresas embebidas en formalina), las técnicas de radiología intervencionista u otras intervenciones presenten la misma efectividad o seguridad que la radioterapia para el control paliativo de las hemorragias vaginales en el cáncer de cuello de útero avanzado. Se necesitan ensayos controlados aleatorizados o estudios comparativos no aleatorizados de buena calidad para determinar la efectividad y la seguridad de estas intervenciones, en comparación con radioterapia, en cuanto al control de los síntomas, la calidad de vida y los efectos secundarios.

Certeza de la evidencia: ningún estudio cumplió los criterios de inclusión, por lo que no hay evidencia certeza adecuada.

Authors' conclusions

Implications for practice

Since the last version of this review, we found no new studies eligible for inclusion. There is no evidence, as we found no comparative studies assessing vaginal packing with or without formaldehyde, tranexamic acid or interventional radiology compared with radiotherapy for women with vaginal bleeding due to advanced cervical cancer. Many non‐comparative observational studies have reported high success rates on the use of these alternative palliative interventions to radiotherapy in the control of vaginal bleeding. However, these are likely to be at critical risk of bias. Clinicians should consider alternative palliative interventions, and counsel women adequately, especially on the challenges of their use in cervical cancer. This is because almost all of the women with advanced cervical cancer need radiotherapy and its use in low‐ and middle‐income countries (LMICs) is limited. Despite LMICs being home to 85% of the world's population, less than 35% of the world's radiotherapy facilities are available in these areas (IAEA 2008; WHO 2011). This inequity goes even further when comparing the availability of radiotherapy services across subregions. This is because effective prevention, early detection and screening services are often also absent, hence, a higher proportion of cervical cancer in these countries is detected at an advanced stage, leaving palliative radiotherapy as one of the only options for treatment (Barton 2006). Thus, the triad of inequity of access, availability and affordability of radiotherapy treatment constitutes a major challenge for the implication of this review in practice, thereby increasing the reliability and desirability of other palliative interventions to improve the satisfactory quality of lives of women with bleeding due to cervical cancer.

In all resource settings, simple local measures, such as vaginal packing and oral haemostatic agents (e.g. tranexamic acid), should be considered prior to, or as an alternative to, more interventional approaches, such as radiotherapy or embolisation. General measures of supportive care should be carried out. The woman's wishes should be respected and distress for women and their families should be minimised.

Implications for research

Currently, there is an absence of evidence for the routine use of vaginal packing (with or without formaldehyde), tranexamic acid, interventional radiology or other interventions compared with radiotherapy in the control of vaginal bleeding in advanced cervical cancer. We encourage the development of clinical trials with random allocation of individual alternative interventions where radiotherapy is unavailable or not suitable due to previous treatment with radiotherapy. Future studies should look at alternatives to radiotherapy rather than a direct comparison with radiotherapy, since comparative randomised controlled trials are unlikely. This is because if radiotherapy is available, it would be used without comparison to other palliative interventions in a trial. However, often in emergency situations, these treatments remain options for consideration, perhaps used in a step‐wise manner.

Background

Description of the condition

This review is an update of a previously published review in the Cochrane Database of Systematic Reviews (2015, Issue 5) on palliative interventions for controlling vaginal bleeding in advanced cervical cancer (Eleje 2015). According to the GLOBOCAN 2018 estimates of cancer incidence and mortality produced by the International Agency for Research on Cancer, cervical cancer ranks fourth for both incidence and mortality of cancer among women worldwide, with only breast cancer, colorectal and lung cancer occurring more frequently (Bhatla 2019; Bray 2018; Ferlay 2019). Worldwide, cervical cancer accounts for estimated 569,847 new diagnoses and 311,365 deaths every year (Bray 2018; Ferlay 2019).

Of the new cases, more than 85% occur in low‐ to middle‐income countries, and in some of these countries cervical cancer is the most common cancer in women (Ikechebelu 2010; Obiechina 2009; Sankaranarayanan 2008). Therefore, nearly 70% of the global burden occurs in the developing world and more than one‐fifth of all new cases are diagnosed in India (GLOBOCAN 2012). The highest regional incidence and mortality rates are seen in Africa, while in relative terms, the rates are 7 to 10 times lower in North America, Australia and New Zealand, and Western Asia (Bray 2018; Ferlay 2019).

Differences are largely affected by the availability of cervical screening. Studies in low‐ to middle‐income countries continue to show a low to moderate level of awareness and utilisation of cervical cancer screening services (Ajayi 1998; Asonganyi 2013; Ayinde 2004; Bradford 2013; Bukar 2012; Gamaoun 2018; Hyacinth 2012; Mukakalisa 2014). This lack of effective screening also explains why in low‐ to middle‐income countries over 75% of affected women present with an advanced stage, whereas in high‐income countries over 75% of affected women present with early‐stage disease (Ikechebelu 2010; NHSCSP 2007‐2010; Umezulike 2007). Despite advances in cervical cancer prevention and diagnosis, outcomes for patients given a diagnosis of advanced and recurrent or metastatic disease are poor, with 5‐year survival rates between 5% and 15% for such cases, and treatment options are very limited (Godoy‐Ortiz 2018; Rosen 2017). Older women, who are at risk for a poorer overall prognosis because of their age, often do not receive the appropriate treatment and are also dying more frequently because of advanced cervical cancer (Quinn 2019).

Early‐stage cancer of the cervix (stage I to IIa; Appendix 1), can be successfully treated by radical surgery (with or without neoadjuvant chemotherapy) or concomitant chemoradiation (FIGO Committee on Gynecologic Oncology 2014; Pecorelli 2009). For advanced cervical cancer (stages IIb to IVb; Appendix 1), the treatment is chemoradiotherapy or palliative chemotherapy with or without radiotherapy (Scatchard 2012). Women with advanced cancer of the cervix often need palliation for distressing symptoms, such as vaginal bleeding. In previous studies, the prevalence of vaginal bleeding in women with advanced cervical cancer ranged from 0.7% to 100% (Adewuyi 2008; Ikechebelu 2010; Shapley 2006; Tarney 2014; Umezulike 2007). Vaginal bleeding is the immediate cause of death in 6% of women with cervical cancer and its management often poses a challenge (Yennurajalingam 2009).

Description of the intervention

Due to the geographical differences in incidence, and with increased rates in low‐ to middle‐income countries, there is a need for different palliative options to treat vaginal bleeding across a variety of healthcare settings. These include: vaginal packing; use of formalin‐soaked vaginal packs; QuikClot combat gauze (Vilardo 2017); cautery; endoscopic haemostatic forceps (Kobara 2015); tranexamic acid; interventional radiology techniques, such as uterine artery embolisation; uterine artery resection; uterine artery ligation and ionising radiation/radiotherapy (Barbera 2010; Fletcher 2002; Pereira 2004; Seider 1988). Potentially helpful radiotherapy approaches include transvaginal orthovoltage treatment (cervical cone irradiation), high‐dose fraction external beam radiotherapy or brachytherapy (irradiation delivered directly into the cervix/vagina). The type and length of treatment will depend on the woman's Karnofsky performance status (a standard way of measuring the ability of people with cancer to perform ordinary tasks). In a woman with a reasonable performance status, stereotactic radiosurgery may be considered to control bleeding (Konski 2005).

In cases of uncontrollable bleeding, uterine artery embolisation may be required. This is a procedure performed using x‐ray imaging for guidance to block feeding blood vessels to the uterus and cervix with metal coils, foam or glue‐like agents. If radiographically directed embolisation is not available, laparotomy with ligation of uterine arteries or anterior divisions of the hypogastric arteries is an alternative, but is likely to be inappropriate in a palliative setting. However, a very few, well‐selected women may derive benefit. Symptomatic anaemia resulting from blood loss can be remedied with blood transfusions once bleeding is stopped (Barbera 2010; Konski 2005).

Since there is no way to guarantee access to immediate radiotherapy in low‐ to middle‐income countries, the use of formalin‐soaked packs has been advocated for the treatment of haemorrhage associated with vaginal bleeding. Other vaginal packing procedures include the use of gauze (Vilardo 2017), lamb's wool or calcium alginate. Monsel's solution (i.e. ferric subsulphate) can also be used to stop vaginal bleeding from cervical cancer (Konski 2005).

Tranexamic acid can be given as an oral tablet or by injection to help to stabilise blood clots and stop bleeding by preventing fibrin degradation. It is effective in controlling blood loss in trauma situations and has been used to prevent blood loss during major surgery (CRASH‐2 Collaborators 2011; Hunt 2015; Ker 2014; Ng 2015).

How the intervention might work

Cervical cancer tends to spread locally before it metastases to distant organs. When the cancer is confined to the pelvis or regional lymph nodes cure is possible with radical hysterectomy, chemoradiotherapy or a combination of the two. In the presence of distant metastatic disease, cervical cancer is generally not curable and treatment is given with palliative intent. In low‐ to middle‐income countries, many women with cancer of the cervix present when the disease is already advanced and metastatic. At this stage, a degree of vaginal bleeding is almost inevitable. If there are distant metastases, there would be limited justification for embarking on aggressive surgical interventions such as bilateral ligation of the hypogastric arteries and so conservative approaches, such as packing of the vagina, use of haemostatic agents and palliative radiotherapy, are more appropriate (Konski 2005).

Oral or intravenous administration of tranexamic acid, three or four times per day, may be used to control/palliate mild‐to‐moderate vaginal bleeding. Tranexamic acid can have both direct and indirect actions. It antagonises the activation of plasminogen (mechanism for dissolving blood clots after they have been formed) by fastening to its kringle enzymatic domain (autonomous protein domains that fold into large loops, which are important in protein–protein interactions with blood coagulation factors). With this action, it reduces the conversion of plasminogen to plasmin. The inhibition of the formation of plasmin is important in controlling vaginal bleeding. In fibrinolysis (fibrin‐splitting), plasmin breaks up the fibrin in blood clots, fibrinogen and other proteins in the blood plasma, such as procoagulant factors V and VIII. In addition, tranexamic acid can antagonise the actions of plasmin directly, though a higher dosage is required to achieve this action than the dosage required to inhibit its action on plasminogen. As well as approaches used to palliate vaginal bleeding, the volume of blood loss needs to be replenished adequately, using plasma expanders or packed red blood cells (CRASH‐2 Collaborators 2011; Mishra 2011).

Tight vaginal packing with simple gauze rolls is a simple first‐aid measure to control haemorrhage. For vaginal packing to be effective, the lithotomy position (legs up in stirrups) is ideal, with the use of an instrument (speculum) to expose the superior part of the vagina aseptically. Sedation or short‐lasting general anaesthesia may be needed. The fornices of the vagina are tightly packed with the aim of maintaining an even pressure in the vagina. While the pack is in situ, the woman requires catheterisation, as she is likely to have difficulty in passing urine, due to compression on the urethra. Other simple measures, such as restricting that woman's mobility, foot‐end elevation of the bed or haemostatic agents (local, oral or parenteral), may enhance the effectiveness of the vaginal packing (Mishra 2011). As long as the vaginal pack is present, broad‐spectrum antibiotics, including metronidazole, may be helpful for treating any underlying infection, which is common with necrotic tumours and can contribute to haemorrhage.

Formalin also acts as a haemostatic agent by initiating a chemical cauterisation, reducing haemorrhage arising from small blood vessels (Chattopadhyay 2010; Vyas 2006). Formalin is an aqueous solution of the chemical compound formaldehyde. The formaldehyde in formalin is responsible for its cauterising property, while the water helps to dilute it, making the solution safer to use. A 4% solution of formalin has no toxic adverse effects (Vyas 2006). This concentration is well tolerated by women, and can help to control vaginal bleeding. Formalin contains noxythiolin (oxymethylenethiourea), tauroline (a condensate of two molecules of the aminosulponic acid taurine with three molecules of formaldehyde), hexamine (hexamethylenetetramine, methenamine), the resins melamine and urea formaldehydes, and imidazolone derivatives such as dantoin. These agents are claimed to be microbicidal, on account of the release of formaldehyde, as well as having some coagulase properties that allow the arrest of bleeding (Adebamowo 2000; Chattopadhyay 2010; Fletcher 2002).

Palliative‐dose radiotherapy can be an effective treatment of vaginal bleeding in cancer of the cervix. The standard curative treatment fraction of radiotherapy is 1.8 Gy to 2.0 Gy in a single fraction. In palliative care, enhanced doses are given with fewer fractions (Mishra 2011).

When vaginal packing fails and radiotherapy is not feasible or available, measures aimed at decreasing the pulse pressure of the blood vessels feeding the cervical tumour can be implemented. These approaches largely comprise surgical interventions or interventional radiology procedures. However, in advanced disease there is often extensive neovascularisation from a variety of feeding vessels, limiting the utility of this approach. Therefore, for meaningful benefit, appropriate patient selection is essential. If excessive bleeding occurs in an end‐of‐life situation, anxiolytics and analgesics to relieve discomfort and distress are important palliative measures, as is psychological and supportive care of the woman and her relatives (Barbera 2010; Mishra 2011).

Mohs' paste works by releasing zinc ions when in contact with a cervical cancer tumour (Yanazume 2013a). Following contact, it precipitates wound proteins to aid haemostasis. These wound proteins include fibrin sealants, thrombin and platelet gels that provide activity during the augmentation or propagation, or both, phase of haemostasis (Glick 2013).

Monsel's solution (ferric subsulphate) is a haemostatic agent that can be applied directly to the bleeding cervix (Davis 1984; Jetmore 1993; Manca 1997; Ratliff 1992; Soyle 1992; Spitzer 1996). It can also be applied in addition to a gauze pack. It works by coagulating proteins, leading to tissue necrosis and eschar formation, enhancing thrombus formation and haemostasis (Glick 2013).

Platinum‐based chemotherapy works by slowing down cancer cell growth and can improve survival and quality of life in advanced disease, with relatively minimal toxicity (Scatchard 2012). However, in cancers, new feeding blood vessels may be abnormal and inadequate leading to lower perfusion of the cancer cells. This may result in a reduced concentration of cytotoxic agents within the tumour. The resulting hypoxia from poor blood supply may lower the proliferating fraction of cancer and reduce the potential cytotoxic effects of platinum‐based chemotherapeutic agents (Lorusso 2014).

Bilateral ligation of the internal iliac arteries (main arteries feeding the pelvis) may work to control intractable bleeding from advanced cervical cancer, especially when vaginal tamponade and blood transfusion are ineffective (Gassibe 1997). However, this requires major surgery and general anaesthetic, which may not be appropriate in a palliative setting.

Endoscopic haemostatic forceps work by soft coagulation which is non‐invasive (Kobara 2015), while QuikClot combat gauze is a synthetic haemostatic dressing used for haemorrhage control (Vilardo 2017).

One further supportive treatment is vitamin K. Vitamin K is essential for the hepatic (liver) production of a number of clotting factors, including factors II, VII, IX and X. Vitamin K treatment may be helpful if there is bleeding in the presence of a derangement of these factors or excessive warfarin therapy (which acts by inhibiting vitamin K‐dependent clotting factor production by the liver) in women with advanced cancer (Shafi 2012).

Why it is important to do this review

Palliative measures to control bleeding in women with advanced cervical cancer are often needed (Barbera 2010; Fletcher 2002; Konski 2005). The intervention chosen is likely to be dependent on the severity of bleeding, but should also take account of effectiveness and the woman's wishes, and those of her family, especially in a palliative or potentially terminal care setting. Radiotherapy may not be suitable (if previously used in the treatment pathway) or may not be readily available, due to limitations in the healthcare setting. However, although there are benefits of non‐invasive or minimally invasive palliative interventions, as compared with more invasive interventions, vaginal packing (non‐invasive intervention) can be painful and distressing and may require sedation or general anaesthesia, whereas other simple measures, such as tranexamic acid, can also have harms associated with them.

There is currently no systematic review that addresses palliative interventions for controlling vaginal bleeding caused by advanced cervical cancer. Systematic reviews on the use of ionising radiation for controlling bleeding show that it can be associated with serious adverse consequences, such as bowel toxicity (Green 2005; Symonds 2004). This damage causes symptoms such as diarrhoea, mucous discharge, cramping, bloating, tenesmus and anal pain or incontinence (Luna‐Pérez 2002). Vaginal bleeding remains a common consequence of advanced cervical cancer. Knowledge of the effectiveness of the interventions to reduce vaginal bleeding in women with advanced cervical cancer is essential to enable evidence‐based clinical decisions. A systematic evaluation of other alternative palliative interventions is therefore warranted to inform decision‐making for this important and distressing health need.

Objectives

To evaluate the efficacy and safety of tranexamic acid, vaginal packing (with or without formalin‐soaked packs), interventional radiology or other interventions compared with radiotherapy for palliative treatment of vaginal bleeding in women with advanced cervical cancer.

Methods

Criteria for considering studies for this review

Types of studies

We searched for randomised controlled trials (RCTs) or controlled clinical trials (CCTs). We only included non‐RCTs that used sensible statistical adjustment in the analysis to reduce the risk of selection bias, especially as women unfit for radiotherapy are given the alternative interventions. This was irrespective of publication status, year of publication or language.

Types of participants

Women (aged 18 years and over) with vaginal bleeding caused by advanced cervical cancer with or without metastasis. We included women with advanced cervical cancer in a palliative setting.

Types of interventions

Vaginal packing (with or without formalin‐soaked packs), tranexamic acid, interventional radiology or other interventions compared with a control group who received radiotherapy.

Types of outcome measures

Primary outcomes

  • Deaths from haemorrhage.

  • Anaemia (defined as haemoglobin concentration less than 11 g/dL).

Secondary outcomes

  • Time to next episode of recurrent vaginal bleeding.

  • Need for subsequent radiotherapy to control bleeding.

  • Need for blood transfusion.

  • Vaginal itching/irritation.

  • Deaths occurring during follow‐up.

  • Serious adverse events (life threatening, resulting in admission to hospital or discontinuation of treatment).

  • Haematological and biochemical adverse effects (e.g. neutropenia, liver toxicity).

  • Vomiting.

  • Anaphylactoid reactions (e.g. dyspnoea, chest tightness, facial flushing, nausea, cyanosis, loss of consciousness, hypotension and death).

  • Other adverse events, including venous thromboembolism and convulsion.

Search methods for identification of studies

Electronic searches

For this update, we searched the following databases up to 21 March 2018:

  • the Cochrane Central Register of Controlled Trials (CENTRAL; 2018, Issue 3) in the Cochrane Library (Appendix 2);

  • MEDLINE via Ovid (March 2015 to March week 2, 2018) (Appendix 3);

  • Embase via Ovid (March 2015 to week 12, 2018) (Appendix 4).

We sought reports in all languages and carried out translations when necessary. We attempted to identify unpublished studies.

Searching other resources

Grey literature

We searched metaRegister, Physicians Data Query, www.controlled-trials.com/rct, www.clinicaltrials.gov, and www.cancer.gov/clinicaltrials for ongoing trials.

Handsearching

We handsearched the following journals from 1980.

  • African Health Sciences.

  • African Journal of Biomedical Research.

  • African Journal of Clinical and Experimental Microbiology.

  • African Journal of Health Sciences.

  • African Journal of Traditional, Complementary and Alternative Medicines.

  • Alexandria Journal of Medicine.

  • Benin Journal of Postgraduate Medicine.

  • Clinics in Mother and Child Health.

  • Continuing Medical Education.

  • Global Journal of Medical Sciences.

  • International Journal of Health Research.

  • Journal of Biomedical Investigation.

  • Journal of College of Medicine.

  • Journal of Ethiopian Medical Practice.

  • Journal of Medical Investigation and Practice.

  • Journal of Medical Laboratory Science.

  • Journal of Medicine and Biomedical Research.

  • Journal of Medicine and Medical Sciences.

  • Journal of Medicine in the Tropics.

  • Journal of Pharmaceutical and Allied Sciences.

  • Journal of Phytomedicine and Therapeutics.

  • Journal of the Eritrean Medical Association.

  • Libyan Journal of Medicine.

  • Malawi Medical Journal.

  • Mary Slessor Journal of Medicine.

  • Medical Journal of Zambia.

  • Nigerian Journal of Parasitology.

  • Obstetrics and Gynaecology Forum.

  • Orient Journal of Medicine.

  • Port Harcourt Medical Journal.

  • Journal of Orthomolecular Medicine.

We also searched the following conference proceedings.

  • American Congress of Obstetrics and Gynaecology.

  • British International Congress of Obstetrics and Gynaecology.

  • British Infection Association conference.

  • Annual Meeting of the International Gynecologic Cancer Society.

  • International Federation of Gynaecology and Obstetrics (FIGO).

Data collection and analysis

Two review authors (GE and AE) independently assessed for inclusion all the potential studies identified from the literature search. We resolved any disagreement through discussion or, if required, we consulted a third review author (GI). If we were uncertain about potential duplicate studies, we corresponded with the authors of the reports. We grouped multiple reports of the same study.

We examined the titles and abstracts to remove obviously irrelevant reports. We retrieved the full text of any potentially relevant reports. We examined the full‐text reports for compliance of the studies with our eligibility criteria. We recorded the reasons for exclusion of studies.

We identified no studies suitable for inclusion in the review.

Results

Description of studies

We identified no studies eligible for inclusion in the review. The studies listed in the Characteristics of excluded studies table were excluded because they were neither randomised trials nor comparative observational studies.

Results of the search

See Prisma flow diagram (Figure 1).


Study flow diagram.

Study flow diagram.

The search strategy identified 1522 references (CENTRAL 30, MEDLINE 265 and Embase 1227). When the search results were merged into Endnote and duplicates removed, there were 1352 unique references. Two reviews authors (GU and AE) independently read the abstracts and articles and excluded 1330 records which did not meet the inclusion criteria. We retrieved 22 full‐text articles and excluded all of the studies for the reasons described below (Excluded studies) and in the Characteristics of excluded studies table. Two review authors (GU and GI) independently searched the grey literature; we found no relevant studies.

Included studies

No studies met our inclusion criteria.

Excluded studies

We excluded 22 studies after obtaining the full‐text papers for the following reasons.

  • Six references were single‐arm (without a comparison group) studies on the use of the intervention radiology in the control of vaginal bleeding in cervical cancer (Banaschak 1985; Ermolov 2003; Ishikawa 1986; Kramer 1999; Mihmanli 2001; Yamashita 1994).

  • Six references were single‐arm (without a comparison group) studies on the use of the palliative radiotherapy in the control of vaginal bleeding in cervical cancer (Biswal 1995; Grigsby 2002; Kim 2013; Kraiphibul 1993; Mishra 2005; Onsrud 2001).

  • One study was presented as a poster abstract at the European Cancer Congress in 2013 in Amsterdam, The Netherlands (Ahmedov 2013). The abstract was a single‐arm (without a comparison group) study that involved the selective embolisation and chemoembolisation (using doxorubicin) of the anterior branch of the internal iliac artery performed on 78 women with cervical cancer.

  • Two studies were case series on the application of formaldehyde‐soaked packs and selective arterial embolisation to stop intractable vaginal bleeding in five women (Fletcher 2002), and three women (Zeghal 2013). None of the women in Fletcher 2002 had cervical cancer.

  • Three references were descriptive reviews that mainly discussed palliative radiotherapy (Konski 2005; Skliarenko 2012), and general treatment options (Pereira 2004), for women with advanced cervical cancer.

  • One study was a single‐arm (without a comparison group) study involving the use of chemotherapy in women with cervical cancer (Adewuyi 2010).

  • One study was a single‐arm (without a comparison group) study on the use of extraperitoneal ligation of the hypogastric arteries in the control of vaginal bleeding in women with cervical cancer (Kwawukume 1996).

  • One study was a single‐arm (without comparison group) study on the use of Mohs' paste to achieved complete haemostasis within a single application in women with cervical cancer (Yanazume 2013a).

  • One study was a short communication on the use of Monsel's solution in controlling vaginal bleeding in cervical cancer (Ratliff 1992).

None of the excluded studies reported on tranexamic acid, which is one of the interventions in the inclusion criteria for this review. For further details of all the excluded studies, see the Characteristics of excluded studies table.

Risk of bias in included studies

No studies met the inclusion criteria.

Effects of interventions

We identified no studies eligible for inclusion in this review. However, we did report results in a narrative discussion in the Agreements and disagreements with other studies or reviews section. These were single‐arm, observational studies and were at a high risk of bias, although we did not objectively assess this, since they did not meet the inclusion criteria.

Discussion

Summary of main results

We identified no studies that evaluated the efficacy and safety of tranexamic acid, vaginal packing (with or without formalin‐soaked packs) or interventional radiology techniques when compared with radiotherapy for palliative treatment of vaginal bleeding in women with cervical cancer. The motivation behind this review was that in low‐ to middle‐income countries, many women with advanced cervical cancers do not have access to radiotherapy treatment for the control of vaginal bleeding. This is because there are few hospitals with radiotherapy centres and those hospitals that do may not be functional due to operative logistics. This makes the use of alternative effective interventions for the control of vaginal bleeding extremely important.

Where a radiotherapy treatment service is available in low‐ to middle‐income countries, many women may not be able to access it due to poverty. Most of the healthcare services available are situated in urban areas and far from the most vulnerable women. The referral system in most of the low‐income countries is generally poor, for reasons including poor road networks, transport infrastructure and a lack of ambulance services. Where radiotherapy services are available, the number of women requiring radiotherapy for all malignancies may be far beyond the capacity of the facilities and expertise available, leading to prolonged waiting times and disease progression. Some cervical tumours are locally advanced at diagnosis and may cause profound haemorrhage, resulting in hypovolaemia and hypotension requiring immediate intervention and blood transfusion when histology results may not be available. Therefore, without histological confirmation in emergency cases, other alternative interventions to radiotherapy are essential to control vaginal bleeding. However, not all women requiring effective treatments for haemorrhage are in a resource‐poor setting and so effective treatments for women in an emergency situation or in whom radiotherapy is no longer indicated (due to previous high‐dose pelvic radiotherapy) are still needed.

Overall completeness and applicability of evidence

We identified no studies eligible for inclusion in this review. However, eligible studies could feasibly be conducted, but would be challenging. The challenge can arise from the difficulty in withholding radiotherapy when available for other non‐radiotherapy interventions, which may appear less effective for the control of bleeding and hence ethically difficult. However, studies could compare different initial treatment options, with palliative radiotherapy held in reserve if the initial treatment was ineffective. The following question remains unanswered: whether tranexamic acid, vaginal packing (with or without formalin‐soaked packs) or interventional radiology techniques are safe and effective when compared with palliative radiotherapy for preventing vaginal bleeding in women with cervical cancer. Although, single‐arm observational studies have shown a potential benefit for the control of vaginal bleeding, using vaginal packing with formalin‐soaked packs or interventional radiology techniques, no definite judgement can be made.

Quality of the evidence

No studies fulfilled the inclusion criteria. Single‐arm, non‐comparator studies formed the only available evidence, but these study designs are at significant risk of bias. Non‐randomised studies are unlikely to address the objectives of the review adequately and, unless well‐designed and adjusted for sensible covariates, are likely to be at high risk of bias and the overall quality of the evidence is likely to remain very low.

Potential biases in the review process

In order to prevent bias in the review process, the Cochrane Gynaecological Cancer Review Group developed and conducted the search. We performed a comprehensive search, including a thorough search of the grey literature and two review authors independently sifted and assessed all studies. We were not restrictive in our inclusion criteria with regards to types of studies as we planned to include good‐quality non‐randomised studies and, we attempted not to overlook any relevant evidence by searching a wide range of quality non‐randomised study designs.

Potentially, the greatest threat affecting the validity of the review is likely to be publication bias. This is because some studies that may have found that vaginal packing alone was not effective in controlling vaginal bleeding and subsequently used the standard intervention such as radiotherapy may not have published it as a comparative study. We were unable to assess this possibility, as we found no studies that met the inclusion criteria.

Agreements and disagreements with other studies or reviews

Currently, there are no systematic reviews of palliative interventions for controlling vaginal bleeding in advanced cervical cancer. We found only one systematic review on palliative radiotherapy for cervical cancer (van Lonkhuijzen 2011), which reviewed optimal palliative radiation doses for the treatment of advanced cervical cancer (van Lonkhuijzen 2011). The review identified eight studies for potential inclusion and none compared the results of different fractionation schemes. Unlike the present review, the van Lonkhuijzen 2011 review did not primarily determine interventions for controlling vaginal bleeding in cervical cancer. Most of the excluded studies were observational retrospective studies, with significant risk of bias. No studies validated endpoints for symptom relief, such as vaginal bleeding, although some reported varying amounts of relief from vaginal bleeding. However, the documentation of acute and late toxicity of radiotherapy was poor (van Lonkhuijzen 2011). The authors concluded that there is a dearth of information to guide the selection of an optimal palliative radiation schedule for treatment of women with advanced cervical cancer, and no evidence to support the common belief that better and longer palliation is achieved with a high doses delivered in multiple, smaller fractions.

However, all the relevant, but excluded, studies appeared to have been designed as single‐arm studies, aiming to assess whether formalin packs, radiotherapy or intervention radiology can be used effectively and in a palliative setting in cases of vaginal bleeding in women with cervical cancer. The study designs allowed no assessment and comparison of the effect of the various interventions on vaginal bleeding control.

For example, in single‐arm studies on the use of radiotherapy in palliative interventions in vaginal bleeding in women with cervical cancers, the overall response rates in terms of control of vaginal bleeding ranged from 90% (Onsrud 2001) to 100% (Biswal 1995; Kraiphibul 1993; Mishra 2005). The treatment was generally well tolerated with a median survival of seven months (Mishra 2005).

However, single‐arm studies (without a comparison group) on the use of interventional radiology (embolisation of iliac or uterine arteries, or both), revealed that the overall response rate in terms of vaginal haemostasis was 70% (Kramer 1999), 93% (Grigsby 2002), and 100% (Ermolov 2003; Mihmanli 2001; Yamashita 1994; Zeghal 2013), though there was documented recurrence of bleeding after initially controlling the bleeding in all women (Yamashita 1994). Interventional radiology has relatively minor acute or long‐term toxicity (Grigsby 2002). In two studies, there were no complications following embolisation therapy (Yanazume 2013a; Zeghal 2013), except for transient fever and lower abdominal pain (Yamashita 1994). Ermolov 2003 concluded that embolisation of uterine arteries is a safe and highly effective alternative to radical surgical interventions in women with acute gynaecological disease complicated by vaginal bleeding. Interventional radiology can provide effective haemostasis and permit either avoidance of surgical intervention or a significant reduction in the volume of intraoperative blood loss. Similarly, Mihmanli 2001 also concluded that transarterial embolisation is a life‐saving procedure in treating intractable vaginal bleeding due to cervical cancer.

Other non‐comparative studies have shown that paraformaldehyde‐soaked vaginal packs could be beneficial (Fletcher 2002). In this intervention, the packs were kept in situ for approximately 24 hours and were removed when there was no further evidence of bleeding. The woman responded well to the treatment and made a full recovery. There was no reported adverse effects (Fletcher 2002).

Mohs' paste achieved complete haemostasis with a single application in one case report (Yanazume 2013a). Haemostasis lasted for more than six months and the woman died of multiple organ failure without further vaginal bleeding (Yanazume 2013a). In another study involving Mohs' paste, eight women with extensive genital bleeding from the uterine cervix or vaginal stump, due to recurrent gynaecological cancer, had Mohs' paste applied directly to the bleeding tumour. The effect of Mohs' paste lasted for at least three months in three women and none of the eight women died of genital bleeding (Yanazume 2013b).

A vaginal pack soaked in Monsel's solution (ferric sulphate, sulphuric and nitric acids) aids haemostasis in women with vaginal bleeding due to cervical cancer (Ratliff 1992), although the majority of its use in the published literature is on premalignant lesions of the cervix (Attarbashi 2007; Tam 2005).

Of seven cases that had extraperitoneal ligation of the internal iliac arteries to control bleeding, five were from women with advanced cervical cancer in a report by Kwawukume 1996. Extraperitoneal ligation of the internal iliac arteries arrested haemorrhage, without the need for any further blood transfusion (Kwawukume 1996). Although the authors of Kwawukume 1996 described the performance of ligation of internal iliac artery abdominally or extraperitoneally, laparoscopic ligation of internal iliac artery has also been described in the published literature (Gassibe 1997; Skret 1994; Skret 1995; Sobiczewski 2002), and so laparoscopy appears to be an alternative procedure to abdominal or extraperitoneal ligation of internal iliac artery in vaginal bleeding due to cervical cancer.

In Adewuyi 2010, where 84/111 women had at least FIGO stage IIIA disease, 81 women had complete cessation of vaginal bleeding, with 69 women having complete cessation on or before the fourth course of cisplatin‐based chemotherapy (ninth week). Adewuyi 2010 concluded that in low‐resource settings, platinum‐based chemotherapy can be used to control vaginal bleeding and improve the quality of life of women pending radiotherapy. Similar to Adewuyi 2010 study, Orang'o 2017 evaluated the effectiveness and feasibility of cisplatinum for palliative treatment of advanced cervical cancer in a resource‐poor setting and concluded it was feasible and led to effective symptom control.

It is worth noting that all of these studies are likely to be at critical risk of bias.

Study flow diagram.

Figuras y tablas -
Figure 1

Study flow diagram.