Treatment for Idiopathic Pulmonary Fibrosis

As everyone working in this field knows, the prognosis of idiopathic pulmonary fibrosis (IPF) is quite dismal (3 years survival of 50%). There have been many treatment regimens that have been suggested, but it is not clear whether any of them are of any use.

The latest issue of the New England Journal of Medicine (NEJM) has 3 articles and one editorial on this subject. The last one on acetyl-cysteine confirms that it does not really work.

However one more phase III trial by Talmadge King’s group on the use of pirfenidone, conducted because the US FDA was not convinced about its efficacy given the conflicting results from previous trials, shows that it does improve decline in lung function and overall survival.

The paper that however does bring a smile and some more hope is the first one by Richeldi Luca et al that shows the usefulness of nintedanib (where do come up with these names) in improving the reduction in FVC in patients with IPF.

There is a lovely short commentary / editorial that puts all these papers in perspective and even addresses the question of whether both nintedanib and pirfenidone should be used together and whether that would help even more or not. This is an interesting thought.

All of these are must reads for all those who have patients with IPF whose care they are responsible for.


PSA Guidelines

This may seem like an unusual post in a radiology blog, but those who do body imaging often get asked about the relevance of PSA levels by patients.

This nice article in the BMJ makes at least one thing clear. If in a > 60-years old male, the PSA is < 1ng/ml, we don’t need to bother.

Arthroscopy versus sham

These are the kind of studies we need more off. Clearly, when the same was done for vertebroplasty versus sham, it raised the hackles of everyone in the radiology world, because two independent studies showed that vertebroplasty was no better than a sham procedure. Of course that controversy in no way seems to have dampened the use of vertebroplasty.

This week’s NEJM has an article that discusses the value of arthroscopic surgery for meniscal degeneration in patients with osteoarthritis and finds the results no better than sham surgery. We already know this intuitively and from anecdotal experience and this is one of the reasons not to overcall meniscal degeneration as tears to prevent unnecessary arthroscopic exploration of frankly osteoarthritic knees.

Thyroid Ultrasound Criteria to Diagnose Possible Malignant Tumors

There are so many important articles that do not get published in the radiology literature.

The recent issue of JAMA Internal Medicine has an article by Smith-Bindman R et al that discusses how to use thyroid USG signs to decide which nodules need to be biopsied.

Three criteria are important – microcalcifications, completely solid appearance and size greater than 2 cm. If we stringently follow the rule that at least two of these criteria have to be present before performing thyroid nodule biopsy, then the sensitivity would be reduced but with a high positive predictive value, but without compromising the ability to pick up malignancy. It is worth reading the article to understand the importance of large population based studies to help us use our radiology signs to better triage patients.


Sub-centimeter Lung Nodule Biopsies

This article from the November issue of the American Journal of Roentgenology discusses the results of 305 CT guided core biopsies / aspirations of small lung nodules, less than 1.0 cm in diameter, over 13 years.

The only major variable that decided success was aspiration instead of core biopsy.

I have been very clear about doing core biopsies for lesions in the body as against FNAC / aspiration for many years now. Tissue today is needed for so much more than just establishing a diagnosis that if we are going to put a needle in the lungs, or other parts of the body, it would be a travesty to come away with only cells, without tissue.

Here is an example of a 7.5 mm lung nodule biopsied earlier this year that turned out to be tuberculosis on histopathology.

Biopsy of 7.5 mm lung nodule

Biopsy of 7.5 mm lung nodule

Does it work!

In a recent issue of JAMA, there is an editorial of whether injection therapy for low back pain works. An evaluation of previous randomised trials and a meta-analysis shows that there is no clinically significant evidence that this works. And yet, we have an entire new field of pain management populated by anesthesiologists in particular but also radiologists and orthopedic surgeons, who specialise in injecting into the epidural space, facets and foramina. A cursory Pubmed review will show a large number of articles on this subject discussing technique and the value of one particular steroid over another….when in reality the efficacy of the very concept of injecting in the spine for pain management is questionable.

This is similar to the vertebroplasty controversy that still really hasn’t died, but for all practical purposes has been subverted by those who practice vertebroplasty so that the procedure can continue to be used across board.

The New York Times last week published a blog based on a recent article in the Mayo Clinic Proceedings that showed how doctors continue to use procedures and therapies that may be ineffective or may even harm patients. This is either due to ignorance or inertia, but in the period that the practice continues, harm continues to be done.

While all of us itch to “do something” each time a patient has a problem, we should constantly be asking ourselves, whether the procedures we perform and the therapies we use, actually work!

Overdiagnosis – Pulmonary Thrombo-Embolism

This one is an eye-opener.

This editorial from British Medical Journal talks about the the issue of incidentalomas, which we as radiologists are aware of – lung nodules, adrenal nodules, breast lesions, etc.

What is less known is the issue of overdiagnosis. And as far the problem with pulmonary thrombo-embolism is concerned, I was quite unaware that a good number of subsegmental lesions do not need treatment, as described in this article in the same issue of the journal.

What does this mean for us? We cannot step down our ability to read thrombo-emboli. However, the treating physicians do need to make a judgement call on whether all of these patients need treatment or not.

The title of this series is “Too Much Medicine” and the article starts with the phrase “When a Test is Too Good..” All very interesting.


The Vanishing Art of Clinical Science – Hyposkilia

This has been written by Dr. Anisha Sawkar Tandon, a consultant radiologist at the St. John’s Medical College, Bengaluru. You can reach her at anisha.sawkar at

I am a practicing radiologist. Over the past few years I have noticed a disturbing trend that threatens to destroy the very nature of medicine. I have been witnessing increasing referrals for radiological investigations even before a clinical history has been taken. We radiologists are familiar with the experience of having to report radiological studies with absolutely the bare minimum clinical history or the complete lack of it. But what has begun to shake me to the core is the fact that even when I have gotten back to the referring clinician for more information, there have been innumerable instances where they simply do not know more just because they have not really talked to the patient in detail, leave alone examined the patient.

Here is an example of one of the situations that I was confronted with when reporting an MRI of the brain of a 58-years old lady who was sent from the emergency department. The referral form mentioned, “Rule out cerebral venous thrombosis (CVT)”. These were the only words of “history” provided. The MRI brain not only had no signs of CVT, but there were some symmetric basal ganglia changes that favored a metabolic encephalopathy, likely related to chronic liver disease. Since the Emergency department is easily accessible with a phone call, I could immediately contact them and ask for more details.

At first, since their question was “rule out CVT”, I attempted to elicit some history related to CVT. The disturbing answer that I got was, “The patient basically has altered sensorium and since CVT is one of the conditions that could cause CVT, we just wanted to rule it out!” I tried to find out more about the altered sensorium and the information I got in response was, “She basically looks like a psychiatric case and also has diabetes and hypertension”. After obtaining this invaluable information, I explained my findings and requested them to assess whether the patient had some form of liver dysfunction since the findings were definitely “metabolic” and not “psychiatric”.

These events had occurred at 6 pm in the evening. The next day, when I decided to follow up on what had happened to my patient, I found this sequence of events that had unfolded after my MRI brain report. The patient, had, at 7 pm undergone an ultrasound of the abdomen, which showed cirrhosis of liver with portal hypertension and ascites and suspicious nodules in the liver. At 8 pm, the patient had undergone a CT scan of the abdomen that confirmed the cirrhosis with regenerative nodules, portal hypertension and ascites. Meanwhile the results of the liver function tests showed findings consistent with chronic liver disease and the patient was known to be HbsAg positive for 5 years. The next morning, a clinical examination by a gastroenterologist and neurologist concluded that the patient was, in fact, in hepatic encephalopathy.

If you analyze this case, it worked out beautifully for the radiologist. What else could a radiologist ask for? An MRI brain, an ultrasound abdomen and a triple phase CT scan of the abdomen in one patient in one evening! What about the patient though? It’s scary to think that not one, but all of the above investigations could have been avoided by just basic history taking and an adequate clinical examination.

I understand that in a busy ER, not every patient with altered sensorium is going to undergo a 30 minutes detailed neurological exam, but a 30 minutes exam wouldn’t be needed if the simple task of talking to the patient had been accomplished which would have brought out the known Hepatitis B status. Whenever I see this kind of situation, I am appalled at what medicine has come to. There are so many more such instances in which an ultrasound of the abdomen is the first time that a large laparotomy scar has been noticed, or an MRI brain study is the first time when a craniotomy has been discovered and the referring clinician has had no knowledge of the same. It’s a different situation when such cases have been directly referred by some remote village primary health center “(PHC) and then perhaps the lack of knowledge by the referring physician can be overlooked. But it’s acutely disturbing when these cases are referred by our own colleagues, specialists in their fields and the first person to touch the patient’s hurting abdomen is a radiologist.

As a radiologist who wishes her field to thrive, I should have no cause for complaint. But as a doctor and a clinical radiologist, I am concerned about the kind of medicine we practice today, where the end result is unnecessary investigations and ultimately a burden on the patient and the system.

I would like to end by quoting Sir Robert Hutchison who said the following words, which sadly are very pertinent today

“From inability to leave well alone; from too much zeal for the new and contempt for what is old; from putting knowledge before wisdom, science before art, and cleverness before common sense, from treating patients as cases, and from making the cure of the disease more grievous than the endurance of the same, Good Lord, deliver us.”

Managing a Pneumothorax – The Fear is Mostly in the Mind

Two weeks ago, in Beirut, I was speaking on the impact of modern imaging in pulmonary medicine, to a group of chest physicians from Egypt. The moment I started speaking on CT guided lung biopsies, a physician got up and started talking about the dreaded risk of pneumothorax. I have had similar reactions in other towns in India, including tier I cities like Delhi.

I explained to him that while a pneumothorax is not uncommon during CT guided biopsies, there are many ways to mitigate its seriousness; single puncture, minimal trauma, bleeding along the tract during withdrawal and arguably puncture side down for a few minutes are measures that reduce the severity and incidence of or prevent the occurrence of pneumothorax.

But even if a pneumothorax does occur, it is a slow event that takes its own time to increase, if it does. If a pneumothorax occurs during the procedure, it can be aspirated at the same time and more often than not, that cares of the problem and we can even continue with the procedure (Figure 1). In case the pneumothorax occurs after we have washed out as a delayed complication, then we can still aspirate it using an over the needle catheter. A recent video describing this procedure has been posted on The New England Journal of Medicine site. While this video shows the technique to be used in patients with spontaneous pneumothorax and describes a blind approach, we can use this simple procedure to aspirate under CT guidance as well and then check the status immediately on the CT scan table.

A and B show the start of a CT guided core biopsy of a subpleural nodule using a coaxial system. Since the nodule is immediately subpleural, the risk of pneumothorax is slightly higher than for a deeper nodule and C shows the developing pneumothorax. In D, the canula has been withdrawn into the pleural space and the pneumothorax was aspirated. E shows the biopsy continuing after the pneumothorax has been aspirated. It did not recur. Five minutes after the procedure, a scan in the supine position (G) shows a thin residual pneumothorax that was non-progressive.

A and B show the start of a CT guided core biopsy of a subpleural nodule using a coaxial system. Since the nodule is immediately subpleural, the risk of pneumothorax is slightly higher than for a deeper nodule and C shows the developing pneumothorax. In D, the canula has been withdrawn into the pleural space and the pneumothorax was aspirated. E shows the biopsy continuing after the pneumothorax has been aspirated. It did not recur. Five minutes after the procedure, a scan in the supine position (G) shows a thin residual pneumothorax that was non-progressive.

In the worst case scenario, an interventional radiologist should be able to put a tube in as well, despite the fact that the need to do this would be very low in experienced hands.

When Procedures Make Intuitive Sense, Despite the Data (or Lack of)…About IVC Filters and Vertebroplasty

Today’s (April 08, 2013) issue of JAMA Internal Medicine brings forth a new controversy regarding another common and popular interventional radiology procedure  – retrievable IVC filter placement.

A review of data in a new retrospective study on the efficacy of IVC filters, has shown that there is only one randomized control trial published in 1998 that showed a reduction in the rate of symptomatic pulmonary embolism in those with filters but with no significant long-term difference in the death rate at 8 years and with an increase in the rate of deep vein thrombosis in those who had filters.

This new retrospective study published in JAMA Internal Medicine concludes, “that the use of IVC filters for prophylaxis and treatment of venous thrombotic events, combined with a low retrieval rate and inconsistent use of anticoagulant therapy, results in suboptimal outcomes due to high rates of venous thromboembolism”

Another study in the same issue discusses the differing rates of use of this technique in California, the main factor being the particular hospital that is providing care. The Society of Interventional Radiology guidelines [4] for the use of an IVC filter list four absolute indications – “recurrent VTE (acute or chronic) despite adequate anticoagulation, contraindication to anticoagulation, complication of anticoagulation, inability to achieve/maintain therapeutic anticoagulation”.  However there is a long list of relative indications with little consensus and that is where there is significant variability, depending on the aggressiveness of the program.

It happened with vertebroplasty when two studies by Kallmes et al and Buchbinder et al, comparing sham procedures with actual procedures showed no significant difference in pain scores between the two groups at all time points. This kicked up a huge storm in the interventional radiology and spine communities, but also forced us to examine our data more objectively and spurred more hospitals and groups to perform their own double-blind control studies, a few of which like the Vertos II trial, have shown that vertebroplasty is better than controlled medical treatment, though there has been no comparison with sham procedures since the 2009 studies.

We all want to believe that “doing” something that is intuitive (cement in an acute fracture, IVC filter stopping downstream emboli) has to help, just like Dr. Zamboni’s belief in treating patients with multiple sclerosis by dilating areas of venous narrowing because of the theory he ascribes to that chronic cerebrospinal venous insufficiency is the cause of multiple sclerosis.

Procedures sometimes quickly become mainstream without data, simply because “they make sense”. Cardiac CT angiography found adopters in patients and celebrities like Oprah Winfrey, simply because it made sense to see the coronary arteries without a catheter, despite the lack of any kind of initial data.

In the end though, we as radiologists need to create compelling stories for the use of techniques, modalities and procedures by making sure that there are enough controlled studies that demonstrate efficacy and safety.

Hopefully in the near to medium future, there will be more date regarding the efficacy (or not) of IVC filters that will clear the air and help patients and the treating doctors make better decisions.

%d bloggers like this: