Tuesday, 28 October 2008

A New Case For October - The final answer

Dear Bloggers

Here are the answers to the latest blog case for October. Thanks for waiting!!


1) Please make a problem list from the above history and physical examination.

  • Abdominal swelling
  • Dyspnoea on exertion and orthopnoea
  • 3 year history RA under control on drug therapy
  • On methotrexate, bucillamine and COX-2 inhibitor
  • History of elevated liver function for 10 years of unknown aetiology
  • JVP elevated
  • Tachypnoea
  • Shifting dullness consistent with ascites
  • Peripheral oedema
  • Percussion dull at both lung bases with decreased tactile vocal fremitus and reduced air entry
  • Fluid containing 77 mmol Na in 500ml with daily rate at 80ml/hour

2) Please list several differential diagnosis and please identify one of the likely contributory causes from the history.

With a history of undiagnosed liver dysfunction, rheumatoid arthritis with use of long-term methotrexate makes me immediately consider drug-induced (MTX) chronic liver disease. However, what goes against this to some extent is the almost complete lack of physical signs of liver disease!

Causes of Liver Cirrhosis (resulting in ascites from portal hypertension) in this patient might include:
  • Drugs e.g. Methotrexate
  • Infective: HBV, HCV
  • Autoimmune: Primary Sclerosing Cholangitis
  • Primary Biliary Cirrhosis
  • Autoimmune Hepatitis
  • Undisclosed alcoholism!
Other GI causes of Ascites might also include:
  • Budd-Chiari Syndrome (associated with autoimmune disease and malignancy e.g. HCC)
  • Portal Vein thrombosis
  • Protein Losing Enteropathy.
Non-GI causes of Ascites include
  • Cardiac: constrictive pericarditis, any cause of right sided heart failure with tricuspid regurgitation e.g COPD, primary pulmonary hypertension, pulmonic stenosis, chronic PE and cardiomyopathy.
  • Renal: Nephrotic syndrome -> drug induced (bucillamine), infective e.g. post-streptococcal, endocarditis, autoimmune -> related to RA (less likely as good control on medication), amyloidosis (chronic inflammation).
  • Ovarian: Ovarian tumour can cause Meig's syndrome producing a transudate!
3) What tests would you do to confirm your hypothesis?

Simple tests first!!
  • Liver function tests and coagulation studies
  • Renal function tests
  • Urinalysis (including protein)
  • Analysis of ascites e.g. serum to ascites albumin gradient, WBC count (ascites portends a high risk for spontaneous bacterial peritonitis), microscopy / culture.
  • Autoantibody studies: Anti-nuclear Abs, Anti-DS DNA Abs, AMA, Anti-SMA, IgA (increased in alcoholics!)
  • Alpha-1 antitrypsin, Caeruloplasmin, Ferritin, alpha foeto protein (AFP)
  • Ultrasound of the liver, kidneys, portal vein, pelvis and heart.
  • Gastroscopy (screening for asymptomatic varices)
4) What would be your evidence based treatment strategy for such a patient?

In this patient, the ultrasound revealed a contracted liver consistent with cirrhosis. However, remember that liver cirrhosis cannot be definitely diagnosed on ultrasound. A liver biopsy is the gold standard method. There was also evidence of portal vein thrombosis.

Cardiac echo showed a normal ejection fraction and no evidence of raised right sided heart pressure.

Renal ultrasound revealed no abnormality.

Lung scintigram was performed to investigate pulmonary fibrosis which was positive with the addition of a mildly raised marker of fibrosis (used in Japan).

Lab studies showed normal AST, ALT, gamma GT and ALP. Bilirubin was 1.8 and PT-INR was 1.34

Urinanalysis was NAD with no protein present.

Autoimmune studies were consistent with rheumatoid arthritis but other immune studies were unrevealing.


All rheumatic drugs were stopped based on the proposition that there was drug induced liver disease on a background of idiopathic hepatic dysfunction plus drug induced pulmonary fibrosis.

Portal Vein Thrombosis: The fact that this patient developed a portal vein thrombosis likely reflects the portal hypertension and venous stasis or reversed venous flow in the portal system. However, infection, malignancy (HCC) or prothrombotic diatheses e.g. Protein C deficiency can promote such this problem. The treatment depends on whether it is considered that such a thrombosis is acute or chronic. Acutely, such thromboses are treated with finbrinolytic therapy and chronically, either with beta-blockers to reduce the formation of varices, surgery, TIPS or anticogulation. Most of you will winge when considering using warfarin in such patients who are at high risk of varix formation and bleeding. However, Condat et al ( Gastroenterology 2001 Feb;120(2):490-7. ) found that anticoagulation in such patients with portal vein thrombosis does not increase the severity of bleeding, with the Authors conclusions being that the risk to benefit of using such treatment in patients with portal vein thrombosis lie in favour of its use.

In decompensated liver disease the intravascular volume is low resulting in the patient having high renin-angiotensin-aldosterone levels. This translates into a situation whereby there is avid salt and water reabsorption such that it is easy to develop ascites. The methodology for reducing ascites should include:
  • High dose spironolactone: 100mg starting dose (UpToDate recommendations) up to 400-600mg /day as a once daily dose.
  • Oral Furoesmide: 40mg once daily. Remember that in Liver disease the absorption of furosemide is unimpaired unlike in chronic heart failure. The reason for this is in heart failure there is generalised oedema of the internal organs from the back pressure of the heart leading to poor absorption. In liver disease the ascites forms from high portal pressure. The reasoning for not initially using furosemide intravenously is due to rapid fluid movement from the intravascular space that can result in worsening renal function. Oral furosemide produces a more gradual reduction in intravascular fluid which can be replaced by the reabsorption of the ascites.
  • Salt restriction: Salt restriction should ideally be no more than 2g/day; this is almost impossible in Japan where the intake of salt can be anywhere up to 12g/day. In order for patients to be losing weight in liver disease, they need to lose >88mmol Na/day. 10mmol is lost via non-renal routes whereas >78mmol needs to lost via the kidneys to produce a net loss. Hence, if a patient is not losing weight it may mean that they are not taking their diuretics and/or they are not salt restricting sufficiently. A simple test is to check a spot Na level to check for compliance.
  • Fluid restriction: Some experts recommend fluid restriction although this must be considered together with parameters such as blood pressure and renal function. There is no one specific maximum amount that can be given to a patient at the outset. It is a trial approach with the aim to work out how much fluid is required to maintain hydration without resulting in formation of ascites or conversely causing renal impairment. However, large volumes of water should be avoided. A general amount of 1-1.5L H20 per day may be acceptable.
  • Daily Weights: The patient should be weighed daily to assess weight loss. In patient with peripheral oedema plus ascites, up to 1.5L per day can safely be lost because fluid is more easily reabsorbed from the peripheral tissues. However, if the patient just has ascites, 0.5L is the generally agreed amount that can safely be lost to avoid hypotension and worsening renal function.
  • OGD: The patient should undergo screening for oesophageal varices and if found, they should be treated accordingly. There are several strategies for reducing portal pressure to reduce rupture of varices and they include the use of beta-blockers and calcium peripheral calcium channel blockers. Remember though, if beta blockers are used in such patients, if they do eventually bleed, you cannot rely on the usual tachycardia to judge about severity of cardiovascular compromise because patients will have a slow pulse (if well beta blocked and compliant !)
  • Avoid Constipation: The patient needs to avoid constipation because this is a precipitant of hepatic encephalopathy and the usual treatment is with lactulose to produce 3 soft stools per day. Remember that lactulose has a significant quanitity of potassium and therefore, K levels need to be checked especially if the patient is using aldosterone antagonists, ACE-Is or ARBs. Conversely, a low K level in liver disease is a precipitant of encephalopathy. In this situation, the kidney produces NH3 resulting in hyperammonaemia. Hence, maintaining a balance of the K level is essential. However, checking the ammonia level is not always clinically helpful. The clinical signs of hepatic encepalopathy e.g. asterixis (liver flap) is a sensitive albeit not entirely specific physical sign of encepahlopathy and is a good guide for improvement in the patient's condition as it will disappear as the encephalopathy dissipates.
  • Biopsy: In patient on methotrexate therapy with abnormal liver function from the onset, a liver biopsy should be performed before starting such treatment. There are several recommendations from various medical societies suggesting at what interval repeat liver biopsy needs to be performed. Essentially, if the patient has abnormal liver function and is started on methotrexate, a liver biopsy needs to be performed before intiating the drug to assess the cause of the liver dysfunction and if deemed permissible to commence such treatment, for every 1g of accumulated dose of the drug, a liver biopsy needs to be repeated. In those patient without liver dysfucntion, a biopsy is recommended after 2g of accumulated methotrexate dose. In such patients, liver dysfunction is more likely to occur after 3-4g of accumulated dose. In patients with existing liver dysfunction the accumulated dose may well be lower.
  • If the patient has ascites, it is not possible to perform a transabdominal biopsy because of the risk of bleeding. However, some centres can perform a liver biopsy internally via the inferior vena cava. The advantage of this is that any bleeding will enter the venous circulation! The disadvantage is that it requires expertise with the inherent risk of pneumothorax, carotid artery puncture, bleeding and infection. However, with the appropriate use of ultrasound guidance to find the internal jugular vein, such risks can be reduced.
As an addition, one other thing to remember in such patient is that they do not need to have a fever to have an infection, they do not need a raised WBC count to have an infection and the CRP should never be relied on especially in liver disease patients (CRP is derived from the liver!!), nor should it in any patient!!!
Patients should be evaluated for infection and a low threshold for treating if such a risk exists. Hence, ruling out infective endocarditis might be worthy in a liver patient with new onset renal failure and a new murmur (this patient did not have a murmur though).

For the purposes of not prolonging the discussion, I will not be discussing the various causes of pulmonary fibrosis. However, in a patient such as this the likely causes are drug-induced and the primary rheumatoid arthritis, although the latter is considered less likely in view of the quiescent nature of the patient's connective tissue symptoms.

Professor Matsumura has kindly answered the case from the history and physical examination.

Thank your for showing me an interesting case again. In this case, the patient has rheumatoid arthritis. It is important to evaluate whether new symptom is related to rheumatoid arthritis or not. I would be careful all patient history, including medications. Dr. Stein had taught me that the importance of evaluation all information.


1) Please make a problem list from the above history and physical examination.

#1 Abdominal swelling and shifting dullness with ascites.
#2 Dyspnoea on mobilizing and no paroxysmal nocturnal dyspnoea (PND)
#3 JVP slight elevation
#4 RR = 30/min
#5 Percussion dull at both lung bases with decreased tactile vocal fremitus
#6 Reduced air entry at both bases.
#7 Pitting oedema of both lower limbs and arms
#8 Swollen hands
#9 Three-year history of rheumatoid arthritis under good control
#10 Taking methotrexate
#11 Taking Bucillamine
#12 Taking COX-2 inhibitor
#13 Containing 77 mmol Na in 500ml with daily rate at 80ml/hour
#14 Elevated liver enzymes over 10 years ago

2) Please list several differential diagnosis and please identify one of the likely contributory causes from the history.

Problem #1 shows ascites.
Problem #2 - #6 disclose massive pleural fluid and possible cardiac failure.
Problem #7 and #8 shows oedema of extremities. Daily sodium administration is 296 mmol, too high from #13. Differential diagnoses are as follows.

Vascular: Less likely Infection:
Liver cirrhosis, non HepB and HepC Neoplastic: HCC, Peritonitis cartinomatosa
Autoimmune: Secondary amyloydosis caused by rheumatoid arthritis, less likely Toxic/Metabolic: Less likely
Trauma/Degenerative: Less likely Iatrogenic:
Nephrotic syndrome caused by Bucillamine
Idiopathic: Less likely
Congenital: Less likely Cardiac failure, liver cirrhosis, nephrotic syndrome, malignancy, or other cause of low serum albumin, for example protein loosing enteropathy can cause ascites.

Possibility of cardiac failure is not high. Because heart sounds is normal, no S3 or S4, no PND, and no crackles in lung sounds. Possibility of liver cirrhosis is low, because no hepatic flap, palmar erythema, spider naevae, and jaundice. I think patient’s serum albumin is low.

In this case, nephrotic syndrome is highly suspected. Because this patient is taking Bucillamine. Most likely diagnosis is nephritic syndrome caused by Bucillamine. Bucillamine use for rheumatoid arthritis is common in Japan. Several cases with nephrotic syndrome caused by Bucillamine were reported in the past. Membranous glomerulonephritis is suspected.

3) What tests would you do to confirm your hypothesis?

I would perform pit recovery time in leg oedema first. If patient’s serum albumin is low, pit recovery time will be more than 40 seconds. This patient’s pit recovery time can be under 40 seconds, if period of leg oedema is short. Urinalysis should be performed immidiately. This is very easy test.

4) What would be your evidence based treatment strategy for such a patient?

Dose of sodium administration must be reduced first. If massive proteinuria is confirmed, Bucillamine should be stopped. In the past prednisolone was prescribed for treatment of this adverse effect in several cases.
Following recent reports suggest that discontinuation of Bucillamine is enough. 1) Obayashi M, et al. Clinical course of bucillamine-induced nephropathy in patients with rheumatoid arthritis. Clin Exp Nephrol 8: 288-290, 2004. 2) Hishino J, et al. Outcome and treatment of bucillamine-induced nephropathy. Nephron Clin Pract 104: c15-19,.2006.

Thank you for such an enlightening opinion and especially about nephritic syndrome associated with Bucillamine in RA patients.

Hello, Dr.B. It's Hirotaka Kato, U of Kumamoto. I email my answer. Please forgive my simple words/email. It is difficult for me this time...I'm looking forward to the answer.

Q1. Problem List
#1. ↑abdominal swelling started 3 m. ago
#2. Dyspnea only on mobilising or lying flat
#3. RA under good control (for 3 years)
#4. Medications (MTX, Bucillamine, folic acid, COX-2 inhibitor)
#5. Elevated liver enzymes pointed out over 10 y. ago
#6. RR 30/min
#7. Percussion stony dull at both lung bases with decreased tactile vocal fremitus
#8. Reduced air entry at both bases
#9. Grossly distended abdomen
#10. Shifting dullness with approximately moderate ascites
#11. Evidence of a previous puncture site on the left lateral abdominal wall
#12. JVP slight elevation
#13. Pitting Edema of both lower limbs
#14. Swollen hands

Q2. Differential diagnosis

I would say the patient's symptoms are caused by congestive heart failure because CHF can unify them. However, lots of negative findings confused me. I was unable to decide the most likely cause. I listed differential diagnosis with no confidence below.
#1. Congestive heart failure due to constrictive pericarditis (?)
# Cor pulmonale due to pulmonary HTN (←RA, MTX)
# Liver cirrhosis
# Nephrosis
# collagen disease (SLE, MCTD, SSc)
# latent malignancy
# Filariasis
# Beriberi

Q3. tests

Chest CT, Cardiac echo Abdominal echo, Abdominal puncture Blood test : CBC(Hb, WBC, Plt) liver enzymes, TP, Aib, BUN, Cr If #1 is suspected, catheter could also be indicated.

Q4. treatment

Firstly, diuretics can be considered for controlling CHF. When CHF cannot be controlled by medications, operation should be performed.

Thank you Kato-san for such a good attempt at answering this difficult case.

In this case, because the physical findings are so non-specific and that it could potentially include three major systems such as cardiac, renal or hepatic, one is somewhat reliant on history to try and elucidate causes. In the end, the laboratory data allows us to check our hypothesis.

The importance of a problem list allows us to list all of the problems and assess each problem individually and then together as a group disorder. It allows us to generate differential diagnoses and plan our investigations logically. Without a differential diagnosis and logical testing, if our only diagnosis is refuted by tests then we have no ability to investigate further. However, by generating the differential diagnosis, when one diagnosis is refuted then the next can be investigated and so on until the real diagnosis (hopefully) is confirmed. That does not mean to say that treatment is witheld until the diagnosis is made. Far from it. All of these conditions are treated in a similar way by reducing fluid and Na. However, many of the above tests can soon be obtained and a streamlined treatment plan can be tailored for the patient.

I hope you emjoyed the challenge of this October case.....another case is coming in November :-)

No comments: