Treatment of Patients with Cirrhosis

Posted by • August 25th, 2016

cirrhosis header

Click to enlarge

In patients with compensated cirrhosis, the 10-year probabilities of ascites, hepatic encephalopathy, and gastrointestinal bleeding are 47%, 28%, and 25%, respectively. These are ominous landmarks; 15% of patients who receive a diagnosis of ascites die within 1 year, and 44% within 5 years. This guide to the practical treatment of patients with cirrhosis summarizes recent developments. It includes advice on medical management, invasive procedures, nutrition, prevention, and strategies to protect the cirrhotic liver from harm. A new Review Article explains in-depth.

cirrhosis table 1

Click to enlarge

Clinical Pearl

• What are some of the medications that should be avoided, and some that may be used, in patients with cirrhosis?

Because of the risk of acute renal failure and gastrointestinal bleeding, nonsteroidal antiinflammatory drugs are contraindicated, except for low-dose aspirin in patients in whom the severity of cardiovascular disease exceeds the severity of cirrhosis. A large, placebo-controlled study involving patients with cirrhosis and ascites showed that although satavaptan alleviated hyponatremia, mortality was higher among patients with recurrent ascites who were receiving satavaptan than among those who were receiving placebo. Because of these findings as well as hepatotoxicity reported with respect to tolvaptan, the use of vaptans in patients with cirrhosis and ascites is not recommended. 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) can be safely started and continued in patients with cirrhosis. Acetaminophen is effective and safe in patients with liver disease, provided that the patient does not drink alcohol.

Clinical Pearl

• When can paracentesis be safely performed in a patient with cirrhosis and coagulopathy?

Paracentesis is relatively safe, even in patients with marked coagulopathy, including an international normalized ratio as high as 8.7 and a platelet count as low as 19,000 per cubic millimeter. It is important not to delay paracentesis in patients with suspected spontaneous bacterial peritonitis. One study showed that diagnostic paracentesis that was performed within 12 hours after the time of first encounter with a physician was associated with increased short-term survival rates. Delayed paracentesis was associated with a risk of death that was 2.7 times as high as the risk associated with early paracentesis. Each hour of delay was associated with a 3.3% increase in in-hospital mortality.

Morning Report Questions

Q: What does the “window hypothesis” postulate about the use of beta-blockers in patients with cirrhosis?

A: Nonselective beta-blockers reduce portal pressures and are used in the primary and secondary prophylaxis of variceal hemorrhage. However, various studies caution the use of beta-blockers in situations such as decompensated cirrhosis with refractory ascites, spontaneous bacterial peritonitis, and severe alcoholic hepatitis. These studies led to the “window hypothesis,” which postulates that beta-blockers are associated with higher rates of survival only within a clinical window. In patients who have early cirrhosis without moderate-to-large varices, beta-blockers do not prevent the development of varices and also result in adverse effects. The clinical window opens when moderate-to-large esophageal varices develop, with or without variceal bleeding, and beta-blockers are indicated for primary and secondary prophylaxis of variceal bleeding. Increasingly, evidence suggests that the clinical window for beta-blockers closes and that they are no longer effective when refractory ascites, hypotension, the hepatorenal syndrome, spontaneous bacterial peritonitis, sepsis, or severe alcoholic hepatitis develops, owing to unfavorable hemodynamic effects in advanced cirrhosis.

cirrhosis fig 3

Click to enlarge

Q: What are some guidelines regarding the use of antihypertensives in patients with cirrhosis?

A: Patients with cirrhosis who have a history of hypertension gradually become normotensive and eventually hypotensive as cirrhosis progresses. Studies of blood pressure in patients with cirrhosis and ascites showed that a mean arterial pressure of 82 mm Hg or less was the single variable that was most strongly correlated with a reduced probability of survival. The probability of survival among patients with a mean arterial pressure of 82 mm Hg or less was 20% at 24 months and 0% at 48 months, as compared with 70% at 24 months and 50% at 48 months among patients with a mean arterial pressure of more than 82 mm Hg. In a similar study, hypotension with a cardiac index below 1.5 liters per minute per square meter of body-surface area predicted the development of the hepatorenal syndrome and decreased survival among patients with cirrhosis and ascites. Because of these hemodynamic changes, antihypertensive agents should be discontinued in patients who have decompensated cirrhosis with ascites or hypotension.

cirrhosis fig 2

Click to enlarge

cirrhosis fig 1

Click to enlarge

A Woman with Back Pain

Posted by • August 25th, 2016

back pain header

Click to enlarge

A 28-year-old woman presented with intractable back pain with radiation to the leg. Examination of biopsy specimens revealed a giant-cell tumor of bone, which accounts for 5% of primary bone tumors and occurs most commonly around the knee. These tumors are usually highly vascular; in some instances it is possible to actually feel the tumor pulsate with each heartbeat. Management decisions were made in a new Case Record.

Clinical Pearl

• Is giant-cell tumor of bone a benign or malignant lesion?

Giant-cell tumor of bone is a benign but locally aggressive neoplasm of mesenchymal mononuclear cells that are thought to have an osteoblast-like phenotype. It was previously thought that the giant cells were composed of collections of mononuclear cells and that both components were neoplastic, but numerous lines of evidence now show that the mononuclear cells are the neoplastic cells and that they recruit large osteoclast-type giant cells that are reactive in nature.

Clinical Pearl

• What are the typical epidemiologic features of a patient with giant-cell tumor of bone?

The typical epidemiologic features of a patient with giant-cell tumor of bone include skeletal maturity, an age between 20 and 45 years, and female sex. Local recurrence is seen in approximately 25% of cases, whereas metastasis occurs in only 1 to 2% of cases and transformation to a high-grade sarcoma in less than 1% of cases. Although metastases usually involve the lung, they may in rare cases involve other distant sites. In contrast to frankly malignant tumors, pulmonary metastases often have an indolent behavior, and thus giant-cell tumor of bone has been considered to be among the rare benign metastasizing tumors.

back pain fig 1

Click to enlarge

back pain fig 3

Click to enlarge

Morning Report Questions

Q: How is giant-cell tumor of bone treated? 

A: In the past, resection with wide surgical margins was the most common treatment for giant-cell tumor of bone, and recurrences were uncommon. However, reconstruction after such procedures was complex and associated with a high rate of complications. Because giant-cell tumor is typically benign, the current standard of care does not include a wide resection unless the tumor is located in an expendable bone (e.g., the clavicle, distal ulna, or proximal fibula) in which resection produces no clinically significant effect on function. Intralesional surgical procedures, which involve curettage of the tumor and débridement of the surface of the healthy bone with a burr, are the mainstay of therapy in most other cases. The local recurrence rates for giant-cell tumor of bone approach 20% among patients who have undergone intralesional surgical procedures; therefore, new local adjuvant and systemic therapies have been sought. Among patients who have undergone an intralesional procedure with cryosurgery, the local recurrence rates are less than 8%, because the freeze–thaw cycle kills cells farther from the burred surface, extending the depth of the effective curettage. However, pathologic fracture and vascular injury have been associated with this form of adjuvant therapy. Other adjuvant therapies, such as therapy with phenol, hydrogen peroxide, or an argon laser, have been used with varying degrees of success, but phenol and an argon laser are not used routinely when the tumor is close to nerves.

Q: Have any new therapies for giant-cell tumor of bone been approved in recent years? 

A: On June 13, 2013, denosumab was approved by the Food and Drug Administration to treat giant-cell tumor of bone. Denosumab is a fully human monoclonal antibody targeting RANKL (receptor activator of nuclear factor-κβ ligand). It was initially developed for the treatment of postmenopausal osteoporosis. RANKL is highly expressed on stromal cells in the tumor that, in turn, release cytokines that support osteoclasts, regulate osteoclastogenesis, and promote progression of giant-cell tumors of bone. A small pilot study involving 37 patients and a phase 2 study involving 282 patients showed that, in patients with either recurrent or unresectable giant-cell tumor of bone, denosumab elicited a tumor response (defined as a partial or complete response according to the RECIST [Response Evaluation Criteria in Solid Tumors] or EORTC [European Organisation for Research and Treatment of Cancer] criteria or the criteria of Choi et al.) in 136 of the 190 patients who could be radiographically evaluated and showed no disease progression in 179 of the 190 patients.

The 70-Gene Signature as an Aid to Treatment Decisions in Early Breast Cancer

Posted by • August 24th, 2016

breast cancer

Click to enlarge

“Your cancer has been successfully removed with surgery, but there may be a role for chemotherapy to protect you in the future.” This message is expressed by oncologists in consulting rooms all over the world. In women with early-stage breast cancer, adjuvant chemotherapy may be offered as an insurance policy against cancer recurrence. Risk of recurrence can be estimated based on algorithms using patient and tumor characteristics, but following these algorithms can lead to overtreatment by exposing some patients to the toxicity of combination chemotherapy, without the benefit.

Genomic testing of tumors may provide greater accuracy in predicting risk of recurrence. However, genomic test results may conflict with existing conventional methods of risk stratification, and it is uncertain whether patients whose clinical and tumor characteristics suggest higher risk of recurrence but whose genomic tumor features suggest lower risk would benefit from adjuvant chemotherapy.

In a study published in this week’s NEJM, Cardoso and colleagues evaluated the role of a genomic testing tool in discordant cases. The study was part of the international multicentre, prospective, randomized-controlled MINDACT trial (Microarray in Node negative and 1 to 3 positive lymph node Disease may Avoid ChemoTherapy). The MINDACT trial compared risk assessment based on the 70-gene-signature assay (MammaPrint®) to assessment with standard clinical-pathological criteria (modified version of Adjuvant!Online) to direct use of adjuvant chemotherapy in 6693 patients with early breast cancer (median age, 55 years); 79% of patients were node negative and 88% had estrogen-receptor/progesterone-receptor–positive breast cancer. Patients were divided into four groups based on clinical-pathological (C) and genomic (G) risk assessments: high C/ high G (n=1806), low C/low G (n=2745), high C/ low G (n=1550) and low C/high G (n=592).

Patients with high C/high G assessments were treated with chemotherapy and those with low C/ low G risk assessments did not receive chemotherapy. The study focussed on the discordant subset of 1550 patients with high C/low G risk. Patients from this group and those with low C/high G risk were randomized to either clinical or genetic risk assignments to determine if they should receive or forego adjuvant chemotherapy. The primary endpoint was to assess whether the distant metastasis-free 5-year survival exceeds 92% in the high C/low G risk group randomized to forgo chemotherapy. Investigators also compared distant metastasis-free 5-year survival rates in patients in the high C/low G risk group who did and did not receive chemotherapy.

In the high C/low G risk group randomized to omit chemotherapy, the primary endpoint was met as the distant metastasis-free 5-year survival was 94.7% (95% confidence interval (CI), 92.5–96.2%). There was a 1.5 percentage point difference in the distant metastasis-free 5-year survival favoring chemotherapy in the high C/low G risk group. However, this secondary analysis was underpowered and not statistically significant (adjusted hazard ratio 0.78, 95% CI: 0.50-1.21, P=0.267).

Of note, among patients with low clinical risk of recurrence, the genetic test provided no added benefit.  Patients with low C had similar outcomes whether they had high or low genetic risk.

In the accompanying editorial, Dr. Clifford A. Hudis and Dr. Maura Dickler consider the significance of a 1.5 percentage-point difference in distant metastasis-free 5-year survival in two women of different ages, priorities, and high clinical risk. They explain, “This difference does not precisely exclude a benefit that clinicians and patients might find meaningful.” The findings of this trial should be helpful in informing decisions about genetic testing of tumors and adjuvant chemotherapy. Despite these technological advances, patients will require careful discussion with their oncologist to determine their next steps.

A Man with Rectal Pain

Posted by • August 18th, 2016

rectal pain headerConditions that have a similar appearance to hemorrhoids include skin tags (which may be tender if they are associated with Crohn’s disease or chronic fissures), condylomata acuminata, condylomata lata, and anal tuberculosis. A 33-year-old man presented with painful bowel movements and rectal bleeding. Physical examination and sigmoidoscopy revealed sentinel skin tags, multiple fissures, and mucosal inflammation of the distal rectum. A diagnosis was made in a new Case Record article.

Clinical Pearl

• What clinical findings suggest that an anal fissure may be due to underlying systemic disease?

In describing anal fissures, the number and location matter. Most fissures are due to mechanical stress; 90% are located in the posterior midline, and 8% in the anterior midline. The presence of fissures in the lateral midline or of multiple fissures makes underlying systemic disease more likely.

Clinical Pearl

• What are the typical symptoms of proctitis?

Proctitis typically causes symptoms such as dyschezia, as well as urgency, tenesmus, and anal discharge. Proctitis is distinct from colitis, which is inflammation of the colon proximal to the rectum. Colitis is usually associated with symptoms such as diarrhea, abdominal pain, bloating, and weight loss. Some conditions cause either proctitis or colitis and others can cause both; the anatomical location of the inflamed region dictates the symptom profile.

Morning Report Questions

Q: What are some of the causes of proctitis that are associated with perianal findings?

A: Perianal abnormalities, including painful skin tags and fissures, are common among patients with Crohn’s disease. Two percent of patients with Crohn’s disease may present with isolated proctitis. Patients with lymphogranuloma venereum (caused by Chlamydia trachomatis serovar L1, L2, or L3) typically present with lymphadenitis. However, lymphogranuloma venereum causes an isolated proctitis without lymphadenitis in 83% of cases that are acquired by anogenital contact, and the secondary stage can mimic Crohn’s disease, with features including perianal fissures, fistulae, and abscesses. Lymphogranuloma venereum is an emerging cause of proctitis in the population of men who have sex with men (MSM) worldwide and is responsible for several outbreaks in Europe in the past 15 years. Although Neisseria gonorrhoeae and C. trachomatis (serovars D through K) are the two most common causes of proctitis among MSM (accounting for 30% and 19% of all infections, respectively), these infections typically do not cause chronic fissures or painful anal papules. Ulcerative colitis and lymphoid follicular proctitis typically do not cause perianal findings.

rectal pain table 1

Click to enlarge

Q: What are some of the reasons that a treatable cause of anal symptoms may be missed?

A: In MSM, symptoms attributable to anorectal infections are often mild or absent, whereas symptoms in the throat or urethra are more severe. Therefore, among MSM, even mild or ambiguous symptoms should prompt comprehensive testing, including urinary nucleic acid testing for gonorrhea and chlamydia and also rectal and pharyngeal culture and nucleic acid testing. Furthermore, MSM should be screened for sexually transmitted infections at regular intervals, even when they have no symptoms. The two most common reasons for missing a treatable cause of anal symptoms are that patients attribute their symptoms to hemorrhoids and that doctors believe them.

Genital Herpes

Posted by • August 18th, 2016

genital herpes table 1

Click to enlarge

Herpes simplex virus type 1 and type 2 cause genital herpes. Antiviral therapy is used for symptomatic outbreaks, and as daily suppressive therapy, it reduces recurrences of symptoms, asymptomatic viral shedding, and the risk of HSV-2 transmission. Only 10 to 25% of patients with serologically confirmed HSV-2 infection are aware that they have genital herpes. Persons with undiagnosed HSV-2 infection are the most important source of new transmissions. A new Clinical Practice article explains.

Clinical Pearl

• In what populations is HSV-1 a more common cause of initial episodes of genital herpes than HSV-2?

Among Americans who are 14 to 19 years of age, the seroprevalence of HSV-1 has decreased by 30% over the past 30 years; thus, an increasing proportion of adolescents lack protective HSV-1 antibodies when they become sexually active. This lack of HSV-1 antibodies has led to an increased frequency of HSV-1 genital herpes acquired from oral–genital sex practices. In some populations (especially young heterosexual women who are 18 to 22 years of age, non-Hispanic whites, and men who have sex with men), HSV-1 is a more common cause of initial episodes of genital herpes than HSV-2.

Clinical Pearl

• What are the clinical manifestations of genital herpes?

The clinical presentation of symptomatic initial genital herpes does not differ between HSV-1 and HSV-2 infection. After an incubation period of 4 to 7 days, multiple lesions appear on the genitals (or adjacent skin), usually bilaterally, and they progress through stages of erythema, papules, short-lived vesicles, painful ulcers, and crusts that resolve over a period of 2 to 3 weeks. Approximately half the patients with symptomatic genital lesions report headache, fever, malaise, dysuria, or tender inguinal lymphadenopathy. However, most patients with initial genital herpes do not have conspicuous lesions and systemic symptoms. In one study, 74% of initial genital herpes infections due to HSV-1 and 63% of initial genital herpes infections due to HSV-2 in women were asymptomatic. Symptomatic recurrences may be preceded by localized prodromal symptoms (e.g., itching or tingling) and are typically less severe than symptomatic initial disease. Lesions are usually unilateral and resolve within 5 to 10 days. Recurrent lesions can be atypical and appear as linear fissures or excoriations. On the basis of clinical examination, it is often impossible to determine definitively whether a first symptomatic episode of genital herpes is initial or recurrent disease, although the presence of a prodrome suggests a recurrence.

Morning Report Questions

Q: What antiviral therapies are used to treat genital herpes?

A: Acyclovir, valacyclovir, and famciclovir are effective therapies for genital herpes caused by HSV-1 or HSV-2 (although most published efficacy data relate to HSV-2). These antiviral medications have excellent safety profiles and rarely cause drug–drug interactions or allergic reactions. Since the efficacy is generally similar, selection of a specific drug is based on the convenience of administration, cost, and clinician preference. Intravenous acyclovir should be used when the manifestations of genital herpes are especially severe or are accompanied by complications, particularly in immunocompromised patients. Topical antiviral therapy for genital herpes is less effective than systemic therapy and is not recommended. Patients with symptomatic recurrences of genital herpes can receive episodic therapy (1 to 5 days of therapy for an acute recurrence) or suppressive therapy (long-term daily drug administration to reduce the frequency of symptomatic recurrences).

Q: What interventions may help reduce the risk of transmission of genital herpes to a susceptible partner?

A: Reactivation of latent HSV infection results in either symptomatic recurrence of genital herpes or asymptomatic viral shedding, which is defined as the presence of infectious HSV on mucosal or skin surfaces that is detectable by means of polymerase-chain-reaction (PCR) assay or culture, without clinical signs or symptoms. Transient episodes (often <24 hours) of asymptomatic shedding of infectious virus from multiple genital sites have been detected with the use of PCR in 80 to 90% of HSV-2–seropositive persons. Shedding occurs on 10 to 20% of days, but at unpredictable intervals. Suppressive therapy significantly reduces the frequency of asymptomatic shedding of HSV-2. In a study involving immunocompetent HSV-2–discordant heterosexual couples, the probability that a partner would acquire genital HSV-2 infection was significantly lower among those who received suppressive therapy with valacyclovir (at a dose of 500 mg once daily for 8 months) than among those who received placebo (14 of 743 susceptible partners with valacyclovir [1.9%] vs. 27 of 741 susceptible partners with placebo [3.6%]; hazard ratio, 0.25%; 95% CI, 0.08 to 0.75; P=0.008). The use of condoms, which reduces the risk of transmission of HSV from an infected partner to a susceptible partner by approximately 30%, should be routinely recommended. The efficacy of administration of antiviral therapy as prophylaxis for an uninfected partner or as postexposure prophylaxis against HSV has not been adequately studied.

Closed-Loop Insulin Delivery during Pregnancy in Women with Type 1 Diabetes

Posted by • August 17th, 2016

closed loop insulin deliveryA 32-year-old woman with type 1 diabetes mellitus presents to your office. She is 8 weeks pregnant. Her blood sugar has been well controlled on a standard insulin pump, but she understands that blood sugar can be more difficult to control during pregnancy.  You have previously discussed the importance of tight glucose control in pregnancy to reduce the risk of complications such as neonatal death, preterm birth, and macrosomia. Your patient has heard about new advances in insulin delivery with closed-loop technology and wants to know if she is a candidate for this technology.

Closed-loop insulin systems use continuous glucose monitors (CGMs) to measure glucose in real time and a computer algorithm to adjust insulin-pump delivery, requiring less input from the patient than a standard insulin pump for daily management of insulin dosing. Studies of this technology in supervised outpatient settings and in the community have shown better glycemic control than with standard insulin pumps, but none of the studies included pregnant women.

A study published in this week’s NEJM addresses the effectiveness and safety of closed-loop insulin delivery in pregnancy. In a 4-week overnight, randomized, crossover study, followed by a 14-week feasibility study, 20 pregnant women (enrolled between 8 and 24 weeks gestation) with a diagnosis of type 1 diabetes and treated with insulin, were randomized to receive insulin treatment with either the closed-loop system (intervention) or a sensor-augmented pump (control), and then crossed over to the other system. After the crossover study and a 2-week washout, patients chose either the closed-loop system or sensor-augmented pump for use day and night through labor, delivery, and the postpartum period (feasibility phase).

The overnight closed-loop system delivered insulin via pump every 12 minutes based on a computer algorithm that calculates insulin doses using CGM. The sensor-augmented pump also uses CGM but requires patient involvement to calculate insulin dosing. Among the 16 patients who completed the crossover study, the primary outcome — the proportion of time with overnight glucose levels within target range (62–140 mg/dL) — was significantly greater during closed-loop treatment than during the control treatment (74.7% vs. 59.5%; P= 0.002). Closed-loop delivery was also associated with significantly lower mean glucose and rate of maternal hyperglycemia. The percentage of time spent hypoglycemic in both groups was low (<2.0%) and did not differ significantly.

During the 14-week feasibility phase, 14 participants chose to continue with the closed-loop device. During this phase, the women using the closed-loop technology had target glucose levels for a mean of 69% of the time; during the 24 hours before delivery, they had target levels for 87% of the time, and continued to have low rates of hypoglycemia.

The investigators conclude that overnight closed-loop therapy resulted in a significant increase in time spent within the glucose target range for pregnancy, without increasing the incidence of hypoglycemia. Despite daily and weekly changes in insulin pharmacokinetics during pregnancy, the system appeared to be safe through the antepartum, labor and delivery, and postpartum periods. Although small, this study provides important data to support the safety of the new closed-loop technology in pregnant women.

Continuing the Discussion on Data Sharing

Posted by • August 16th, 2016

The recent publication of four Perspective articles on data sharing prompted vigorous discussion on Twitter and on

The Journal is committed to data sharing in the setting of clinical trials, and we are supportive of an open discussion around this critical issue. We will continue to create opportunities for this dialogue.


Raynaud’s Phenomenon

Posted by • August 11th, 2016

raynaud cover picAlthough laboratory testing provides important information about the hemodynamic and physiological features of Raynaud’s phenomenon, clinical assessment by means of history or direct observation remains the best approach for diagnosis. Most experts agree that at least biphasic (white [pallor] and blue [cyanosis]) change in the skin color of the digits is needed. A new Review Article updates the understanding of the pathogenesis, the approach to management, and current approaches to drug therapy.

Clinical Pearl

What type of disease is most often associated with secondary Raynaud’s phenomenon?

Patients who initially present with Raynaud’s phenomenon and then have progression to an underlying secondary disease generally have a connective-tissue disease, commonly systemic sclerosis (scleroderma). One study showed that 37.2% of 3029 persons who were thought to have primary Raynaud’s phenomenon subsequently had a connective-tissue disease.

Clinical Pearl

Are there factors which help to predict whether a patient with isolated Raynaud’s phenomenon will go on to manifest a secondary underlying disease?

Recent studies have emphasized that factors such as the onset of Raynaud’s phenomenon near the age of 40 years, severe frequent events, and the presence of abnormal nailfold capillaries can help predict whether a connective-tissue disease will develop and are especially helpful in identifying early scleroderma. A survey that followed 299 patients with primary Raynaud’s phenomenon for a median of 4 years showed that if capillaroscopy reveals normal nailfold capillaries and if all tests for scleroderma-specific antibodies are negative, then the chance that scleroderma will develop is less than 2%.

raynaud fig 1 

Morning Report Questions

Q: What nonpharmacologic therapies are recommended for Raynaud’s phenomenon?

A: Although evidence from clinical trials is still needed to provide solid guidelines, there is little doubt that effective cold avoidance and stress reduction constitute the foundation of any treatment program for Raynaud’s phenomenon. This approach alone treats the majority of patients who present with primary Raynaud’s phenomenon and is also a major factor in treating patients with secondary Raynaud’s phenomenon. Systemic warming is best accomplished by keeping the whole body warm with layered clothing, gloves, and head covering; avoiding rapidly shifting temperatures, such as rushing into an air-conditioned area; and avoiding cold and breezy conditions. Local hand warming with gloves and rubbing the hands in warm water or with chemical warmers can help prevent an attack or speed recovery. A variety of factors can potentially aggravate the disorder and should be avoided, including smoking and the use of sympathomimetic drugs, agents for the treatment of attention deficit–hyperactivity disorder, and agents for the treatment of migraine headaches.

Q: If nonpharmacologic approaches do not control Raynaud’s phenomenon, are there effective pharmacologic therapies? 

A: Drug therapy is initiated when nonpharmacologic approaches are ineffective in reducing the severity of vasospastic attacks and improving quality of life. Currently, a popular clinical practice is to use a long-acting dihydropyridine calcium-channel blocker as monotherapy, adjusted to the maximally effective dose with the fewest side effects. A 2005 meta-analysis of randomized trials involving 361 patients with primary Raynaud’s phenomenon showed benefit with the use of calcium-channel blockers, with a reduction in the frequency of attacks by an average of 2.8 to 5 attacks per week. A recent Cochrane review provided moderate-quality evidence that oral calcium-channel blockers are minimally effective in the treatment of primary Raynaud’s phenomenon as measured by the frequency of attacks; there were 1.72 fewer attacks per week (95% confidence interval [CI], 0.60 to 2.84) with a calcium-channel blocker than with placebo. A 2001 meta-analysis of randomized trials involving patients with scleroderma and Raynaud’s phenomenon supported the view that these drugs are moderately effective in patients with secondary Raynaud’s phenomenon. If calcium-channel blockers are ineffective as determined by self-reported responses by patients on an office-administered Raynaud’s Condition Score, if they cannot be taken because of side effects, or if there is persistence of a secondary complication with digital ischemic lesions, popular options include the use of a phosphodiesterase type 5 (PDE-5) inhibitor or a topical nitrate, alone or in combination with the calcium-channel blocker. There is also some evidence to support the use of selective serotonin reuptake inhibitors (SSRIs) or angiotensin II–receptor blockers (ARBs).

raynaud fig 2raynaud fig 3raynaud table 1

Thymectomy in Myasthenia Gravis

Posted by • August 11th, 2016

myasthenia cover pic Wolfe et al. conducted the Thymectomy Trial in Non-Thymomatous Myasthenia Gravis Patients Receiving Prednisone Therapy (MGTX), an international, randomized, single-blind (rater-blinded) trial, to determine whether extended transsternal thymectomy combined with a standardized prednisone protocol would be superior to prednisone alone after 3 years, with respect to lessening myasthenic weakness, lowering the total dose of prednisone, and enhancing quality of life.

In patients with nonthymomatous myasthenia gravis, thymectomy plus prednisone was associated with better clinical outcomes than prednisone alone. Patients treated with thymectomy had fewer hospitalizations for exacerbations and required lower prednisone doses. A new Original Article explains.

Clinical Pearl

• What has been the rationale for performing thymectomy in patients with myasthenia gravis?

Thymectomy for the treatment of myasthenia gravis is based on several lines of evidence that support a central role of the thymus in the pathogenesis of the disease. Thymomas are present in 10% of patients with myasthenia gravis, and thymectomy is considered to be mandatory to prevent further spread. Up to 70% of the remaining patients with myasthenia gravis have hyperplastic thymic changes that are not seen in healthy persons. However, the success of immunotherapy has raised questions regarding whether such an operation is necessary.

Clinical Pearl

• What is known about the possible benefits of thymectomy in myasthenia gravis?

Thymectomy has been a mainstay in the treatment of myasthenia gravis, but there is no conclusive evidence of its benefit. The first reported use of thymectomy in patients with nonthymomatous myasthenia gravis was 75 years ago. Of six patients who underwent surgery, three had a favorable response. Subsequent retrospective studies have shown benefits of thymectomy in patients with nonthymomatous myasthenia gravis but with widely varying rates of clinical improvement or remission. A compilation of retrospective studies comparing surgery with medical management did not show a difference in remission rates. Two studies that showed clinical improvements after thymectomy indicated that benefit occurred in the first few years after the procedure, but after 5 years, rates of clinical improvement were similar among surgically treated patients and those who were treated medically. Observational studies have not shown benefits of thymectomy, perhaps because of the effectiveness of modern immunotherapeutic approaches. Despite calls for a randomized, controlled study, data are lacking, and uncertainty persists regarding the benefit of thymectomy and the clinical characteristics of the patients who should be offered the procedure.

Morning Report Questions

Q: Did the MGTX trial provide evidence supporting the role of thymectomy in myasthenia gravis?

A: The MGTX trial provides evidence supporting the use of thymectomy for improving clinical outcomes and reducing the need for immunosuppressive therapy in patients with myasthenia gravis. Over a period of 3 years, thymectomy was associated with a more favorable outcome than was prednisone alone, with a difference in Quantitative Myasthenia Gravis scores that correlated with clinically significant improvement. The time-weighted average required alternate-day prednisone dose was significantly lower in the thymectomy group than in the prednisone-only group.

Q: What other benefits were associated with thymectomy in the MGTX trial?

A: In the MGTX trial, fewer patients were hospitalized for exacerbations of myasthenia gravis in the thymectomy group than in the prednisone-only group (9% vs. 37%, P<0.001). Similarly, other findings favored thymectomy over prednisone alone, including the time-weighted average score on the Myasthenia Gravis Activities of Daily Living scale (2.24 vs. 3.41, P=0.008), azathioprine use (17% vs. 48% of participants, P<0.001), and the percentage of patients who had minimal-manifestation status at month 36 (67% vs. 47%, P=0.03). The survey regarding treatment-associated complications showed no significant difference between the two treatment groups over a period of 3 years (P=0.73).

myasthenia fig 1myasthenia table 3

Randomized Trial of Thymectomy in Myasthenia Gravis

Posted by • August 10th, 2016

As a third-year medical student in my neurology clerkship, I went to the emergency room to evaluate Ms. B, who presented with eyelid weakness and fatigue. She was a 45-year-old woman who was diagnosed 2 years earlier with myasthenia gravis (MG). This was her third hospitalization for exacerbation of the disease. At the time of diagnosis, the CT scan showed no evidence of thymoma. She had been taking pyridostigmine and prednisone, but had steroid-related side effects, including weight gain and irritability. Because of the severity of her symptoms, we treated Ms. B with IVIG and discharged her from the hospital several days later. I knew this wouldn’t be the last exacerbation for Ms. B. In the 10% of patients with MG who have a thymoma, thymectomy is the standard of care. But would this patient benefit from surgery?  Six years later, we now have more information to help answer this question.

Many patients with non-thymomatous MG have hyperplastic thymic changes that are not seen in otherwise healthy individuals, leading to speculation about the role of thymectomy, even in the absence of thymoma.  Several retrospective observational trials have yielded mixed results on rates of improvement and duration of response associated with thymectomy as a treatment for non-thymomatous MG. Based on these limited data, thymectomy is often performed in some patients with non-thymomatous MG with generalized disease and positive acetylcholine receptor antibodies. However, the benefit of thymectomy for such patients had never been examined in a randomized trial.

A multi-center, international, randomized-controlled trial, conducted by the MGTX Study Group and published in NEJM this week, aims to answer this question.  The trial enrolled 126 patients, from 67 centers, with recently diagnosed myasthenia gravis and at least mild generalized disease. Patients were randomized to undergo transsternal thymectomy plus prednisone or prednisone alone. The investigators developed an alternate-day prednisone dosing protocol with specific thresholds to adjust doses based on symptoms.  If patients did not achieve minimum manifestation status (defined as no symptoms or functional limitations from myasthenia gravis) or had intolerable prednisone-related side effects, they could receive azathioprine. The dual primary outcome included a validated quantitative myasthenia gravis score and total prednisone exposure to ensure that any differences in outcomes were attributable to thymectomy and not to higher doses of glucocorticoids.

Analysis of the primary outcome revealed that during the 36-month follow-up, patients who underwent thymectomy had improvement in quantitative myasthenia gravis scores, compared with patients who did not (6.15 vs. 8.99; P<0.001), and received a significantly lower prednisone dose (average alternate-day dosage, 44 mg vs. 60 mg; 95% CI, 7-25 mg, P<0.001). The thymectomy group also had lower hospitalization rates (9% vs. 37%, P<0.001), a smaller percentage of patients who received azathioprine (17% vs. 48%, P<0.001), and a higher proportion of patients who achieved minimal manifestation status (67% vs. 47%, P=0.03). Treatment-associated complications did not differ significantly between the two groups. Study limitations include the unblinded design (due to the unethical nature of performing sham surgery) and the necessary, but imperfect ,metric of pill counts to calculate prednisone exposure.

In an accompanying editorial, Dr. Allan Ropper, Deputy Editor of NEJM observed that the absolute difference between the two groups in the primary outcome was modest, but reported secondary outcomes were of high importance to patients.  Dr. Roper notes that “[t]he data from the trial will be useful in counseling patients but they offer no further clarity regarding selection of patients for thymectomy because subgroup analyses did not allow conclusions regarding the effects of surgery for men vs. women or age under vs. over 55 years.”

Overall, this trial used the rigor of a randomized-controlled trial to answer an important, decades-old question: Do patients with myasthenia gravis without thymoma benefit from thymectomy? The trial demonstrates that thymectomy can lead to improved outcomes without significant side effects, and adds important data to this field for clinicians, investigators, and most importantly for patients with myasthenia gravis.