Radiologists meet with heavy collateral damage
was at a large brasserie in Paris a few weeks ago. Looking for the toilets, I passed by a blackboard with information for the waiting staff. The advice for the day: "Push the fresh fish!"
The waiters did. I ordered steak.
Nearly one-third of the world's fish consumption stems from offshore farms, mostly in bays and estuaries along the sea coasts. In Europe, industrial salmon farming used to be a Scandinavian business. Nowadays, a Norwegian company dominates the world salmon farming industry, from Norway to Scotland, Canada, and Chile. Fish farming has grown into a sophisticated industry that aims to profit from scientific and technological advances.
I was first exposed to aquaculture when a research team at our institute was contracted to perform high-resolution MR spectroscopy and imaging on salmon. The salmon farmers were disturbed to find that live fish transported in tanks were losing their taste. MRS revealed changes to the chemical composition of the meat, changes that were blamed on the fish being stressed. It was suggested that valium added to the water could calm the fish, prevent these changes, and make the salmon tastier.
I do not know whether the producers ever implemented this solution. Salmon farmers claim not to use hormones, but they do employ fungicides, pigments to artificially turn the white flesh of farmed fish wild salmon red, and antibiotics. The dosages stay within legal limits that vary from country to country . The industry had been criticized for their ecological effects on the environment, and a risk-benefit analysis has suggested that consumers should not eat farmed fish from Scotland, Norway, or eastern Canada more than three times a year (!) to minimize the possible health hazard .
Risk-benefit management is part of the curriculum at all business schools. In the case of fish farming, it means the best shareholder value for the best possible quality and highest quantity of salmon. The question is how many chemicals and antibiotics, if any, are needed to reach this objective.
A similar situation holds for the healthcare industry. During the past 30 years, there have been major developments in medical imaging. Many have been transformed from research into medical reality through major investment by companies. These developments often carried major business risks. Other risks, deemed minimal, were sometimes deliberately overlooked.
Risks were sometimes deliberately overlooked.
Warning voices were not heard.
In late summer 1988, exactly 20 years ago, I saw a poster at a major scientific meeting in Berlin. I still remember it – hanging close to one of the big windows in a corridor of the conference hotel.
The paper was about the incorporation of gadolinium into bone, a topic of limited interest to most congress participants because the newly developed gadolinium contrast agents were known to be stable and safe despite the toxicity of their prime component, gadolinium. Free gadolinium is deposited in the liver, bones, skin, and lymph nodes. Therefore it has to be tightly held by a claw, a chelate. These chelates characterize and differentiate the various contrast agents from each other. Basically, they come in two configurations: linear, as for instance in Magnevist and Omniscan; and cyclic, as in Dotarem or Prohance.
Although they are considered unspecific contrast agents, the target organ of these compounds is the kidney, and they require proper renal function to be excreted. Any problem with the kidneys will leave residues in the body.
The stability of the different gadolinium chelates in the body varies drastically. All release gadolinium ions, but some substantially more than others. As a precaution, additional chelate is part of the contrast agent mixture injected – to try to catch the free gadolinium in the patients' bloodstream. Linear agent O dissociates in 30 seconds and releases the gadolinium ion, linear agent M in 10 minutes, and cyclic agent P in three hours ; cyclic agent D remains undissociated for many hours. It would be advisable to use agent D in patients whose kidney function is impaired. Unfortunately, the latter agent was and is available only in Europe, not in the U.S. because somebody there blocks the patent.
Rumor has it that one of the companies with a linear agent had the choice: linear or cyclic. They went for linear, most likely because they could market it faster at lower cost.
Magnevist was the first agent on the market, in 1988. It was followed by Dotarem the year after, Prohance in 1992, Omniscan in 1993, and Optimark in 1999. To date, more than 200 million examinations with such agents have been performed worldwide.
By the late 1980s, all relevant details about these contrast agents were included in our MR imaging teaching courses. Special courses for employees of pharmaceutical companies were available, too. I received a call from one manufacturer following such a course, complaining that I had stated that their compound had led to fatal reactions. I had never made any such statement. I had explained that any injection, be it of a contrast agent or water, may have immediate, acute side effects, for instance, anaphylactic shock.
With hindsight, I wonder whether they were afraid of something – because it had been pointed out as early as in 1988 at a conference in Norway that a macrocycle approach to contrast agents would create inert gadolinium complexes . At that time, no manufacturer ever spoke about the possibility of unknown or unexpected late effects, although the drama surrounding the x-ray contrast agent Thorotrast should have been well remembered .
The years passed, and gadolinium- based contrast became increasingly popular. The indications expanded from head and spine imaging to body applications, and then to pediatric examinations. Radiologists started looking into off-label indications as well. The moves in that direction started early, with researchers first playing with double and triple doses, then turning to MR angiography.
Time-of-flight and phase-contrast techniques without the use of a contrast agent had not proved better than x-ray techniques. The angiography contrast agents in the pipeline were not yet ready. So why not use the existing agents at a higher dose? Techniques were patented by MRA protagonists, and companies were easily persuaded to give a helping hand to the off-label application.
Warning voices were not heard. Gadolinium-based agents were believed to be risk-free. The physicians involved had no background in the complex mechanisms of the behavior and biochemistry of gadolinium contrast agents. Laboratory chemists and biologists did not understand the paths of medical thinking. Marketing staff were deeply ignorant, and those in R&D were too low down the pecking order to be asked.
Meanwhile, a big scare broke loose on a different front. A wave of company- sponsored satellite meetings flooded the free-lunch floors of scientific conferences, each bringing the message that only certain iodinated contrast agents should be used in patients with restricted kidney function. The entire move was more a marketing operation than anything based on scientific evidence. Its aim was most likely to prevent harmed patients from further harm. But the road to hell is plastered with good intentions.
Some people thought further. Others smelled money. Why not replace x-ray examinations in kidney patients with contrast-enhanced MR imaging? These examinations would involve no ionizing radiation and would use less aggressive contrast agents. Studies by radiologists performing MRA had shown that this practice was safe. They all focused upon nephrotoxicity, stating that high-dose gadolinium chelates were significantly less nephrotoxic than iodinated contrast agents .
To corroborate these results, a phase III clinical study was performed in a major university hospital. Several dozen people with renal failure were enrolled. The trial ended in disaster, though not immediately. At the end of the trial, everything looked positive . As in the Thorotrast case, the damage became clear later, after weeks, months, years.
Many of the trial's participants developed strange symptoms. They were identified as having developed nephrogenic systemic fibrosis, a systemic disorder characterized by thickening and tightening of the skin and subcutaneous tissues. NSF can include fibrosis of skeletal muscle, lung, liver, testes, or myocardium. The effects are irreversible–believed to be related to gadolinium freed into the body. It is an iatrogenic disease with a possibly fatal outcome .
NSF was originally known as nephrogenic fibrosing dermopathy owing to the typical initial symptoms of symmetric swelling, discoloration, and pain of the lower legs. The first case was described nine years after the introduction of Magnevist, four years after the introduction of Omniscan. The first major article describing 15 patients with NSF appeared in 2000 .
The accepted school of thought today is that gadolinium-based contrast agents administered in high doses have high nephrotoxicity . The medical community also seems to have realized that free gadolinium may accumulate in tissues when contrast is administered in high doses or in repeated examinations to patients with severe kidney disease. The number of new NSF cases seems to be on the decline now that this information has been taken on board.
Two studies just published reported incidence of NSF between one in 2913 and one in 44,224 patients, depending on contrast agent, dose, frequency of injection, and severity of kidney disease [11, 12]. These figures make me wonder. Basically, these numbers don't say anything because the selected groups are not comparable. The rates are quite high, considering that only around 250 cases have been reported worldwide. These cases are accompanied by at least 500 more or less learned papers on the topic since 2000.
In the general patient population, the overall rate of incidence of NSF is probably one or two orders of magnitude lower, and the chance of an average patient being affected after undergoing contrast-enhanced MR imaging is negligible. The agents' diagnostic benefit is undisputed – if used properly, not in doubtful indications. Contrast agents are used for more than 50% of all MR imaging examinations in some countries. There is no medical reason for that.
In the words of Dr. Peter Marckmann from the department of nephrology at Copenhagen University Hospital, "Unfortunately, there is no proven curative treatment. It is therefore essential that future cases of nephrogenic systemic fibrosis are prevented." 
There is an outrage among radiologists. At every minor or major radiological meeting, entire sessions are devoted to NSF, navigating between Scylla and Charybdis. Everybody is looking for explanations and somebody to blame.
So who or what is to blame? Greed is the most likely culprit. It is not only stupidity, "negligence," or "human error"– those unavoidable factors cited at news conferences: "Sorry, it happened – it was unpredictable, an accident." A friendly smile and back to today's agenda.
Frankly, the first to blame are those radiologists who used the drugs off-label. If you apply a drug outside the recommended and approved protocol, you bear the responsibility. It is as simple as that.
If you talk to fellow radiologists, you will soon discover that most do not know how these drugs function; "gado" is no "dye." There is a rather complicated mechanism behind these drugs' contrast altering actions. Furthermore, even the most mentally challenged physician knows that the pharmaceutical industry will not inform you about possible unpleasant characteristics of their products, particularly if they hope these characteristics will not show up if the drug is applied within the strict limits of its approved use.
But the fingers itch. Let's make more money even if there is a rotten smell: "Push the fresh fish."
I reckon that the main factor is ignorance and lack of ethics. Radiologists should know what they inject. This is something one does not necessarily learn at free-lunch or free-dinner meetings. If you pay for yourself and try to understand the boring background, then you buy peace of mind. It's you, the radiologist, who will have to face the patients afterwards.
And the companies? Those answerable have long moved on, gliding to new sinecures on a golden parachute. By the way, two top managers at a multinational fish farming company used to be top managers at one of the contrast agent producers.
1. Barrionuevo A. Salmon virus indicts Chile’s fishing methods. New York Times (online ed.) 22 March 2008. Ed. note. NYT (online ed.) 13 May 2008.
2. Foran JA, Good DH, Carpenter DO, et al. Quantitative analysis of the benefits and risks of consuming farmed and wild salmon. J Nutr 2005; 135: 2639-2643.
3. Penfield JG, Reilly RF. What nephrologists need to know about gadolinium. Nat Clin Pract Nephrol 2007; 3: 654-668.
4. Tweedle MF. Work in progress toward nonionic macrocyclic gadolinium (III) complexes. in: Rinck PA (ed.). Contrast and contrast agents in magnetic resonance imaging. Trondheim and Mons: The European Workshop on Magnetic Resonance in Medicine (EMRF). 1989. 65-73.
5. Dahlgren S. Tumors following administration of Thorotrast. A review of the literature. Radiographica 1962; 4: 87-94.
6. Prince MR, Arnoldus C, Frisoli JK. Nephrotoxicity of high-dose gadolinium compared with iodinated contrast. J Magn Reson Imaging 1996; 6: 162-166.
7. Joffe P, Thomsen HS, Meusel M. Pharmacokinetics of gadodiamide injection in patients with severe renal insufficiency and patients undergoing hemodialysis or continuous ambulatory peritoneal dialysis. Acad Radiol 1998; 5: 491-502.
8. Marckmann P, Skov L, Rossen K, Thomsen HS. Clinical manifestation of gadodiamide-related nephrogenic systemic fibrosis. Clin Nephrol 2008; 69: 161-168.
9. Cowper SE, Robin HS, Steinberg SM, et al. Scleromyxoedema-like cutaneous diseases in renal-dialysis patients. Lancet 2000; 356 (9234):1000-1001.
10. Nyman U, Elmståhl B, Leander P, et al. Are gadolinium-based contrast media really safer than iodinated media for digital subtraction angiography in patients with azotemia? Radiology 2002; 223: 311-318.
11. Wertman Ba R, Altun E, Martin DR, et al. Risk of nephrogenic systemic fibrosis: evaluation of gadolinium chelate contrast agents at four American universities. Radiology 2008; 248: 799-806.
12. Prince MR. Zhang H. Morris M, et al. Incidence of nephrogenic systemic fibrosis at two large medical centers. Radiology 2008; 248: 807-816.
13. Marckmann P. An epidemic outbreak of nephrogenic systemic fibrosis in a Danish hospital. Europ J Radiol 2008; 66: 187-190.