Tuesday, 20 May 2014

MERS-CoV detections: The April wave recedes...

So welcome to the 114th Week of MERS-CoV among us. That week numbering may change shortly. Stay tuned if week numbering is your thing.

We currently have a tally of 649 detections of MERS-CoV or viral antibodies in humans. I don't list camel numbers. My count says 192 fatalities among infected people, resulting in a proportion of fatal cases of 29.6%. That seems high. Because, until very recently, the Kingdom of Saudi Arabia's Ministry of Health did not regularly report deaths alongside their date of illness onset, it has been an interesting hobby to try and link them. The number is solid so along as the MOH has not been doubling up in the reporting or coming back later to re-report deaths. You'll be familiar with these issues if you follow me on Twitter.

I made a point of saying antibodies earlier because I am going to be including these sorts of laboratory data in my tally when produced by trustworthy laboratories who have described their methods and shown some validation data and an understanding of what the cross-reaction issues are when dealing with MERS-CoV serology. This will be despite the current WHO MERS case definition not allowing for inclusion of people who only have antibody but no virus or viral RNA detected in their samples. There may be some hiccups with MERS-CoV antibody testing along the way, but we need these data in humans and it's good to see the wheels rolling on this at last.
[One of those hiccups occurred 28-May-2014, when the test result from an Illinois man who had originally tested positive in an Ab test, was retracted.]

In my estimation though, serology (the testing of human sera for antibodies against a virus here, the main target being IgG which takes a couple of weeks to become detectable after infection) is a much more reliable way of defining an infection by MERS-CoV virus than by relying on patient recall bias of symptoms 2-weeks ago, or from directly observing signs and symptoms that are nondescript and difficult to distinguish, alone. The latter approach has been the mainstay of identifying cases of human infection for a very long time; still is. This approach is especially important during times of outbreak and pandemic when labs are swamped by testing requests and it must be assumed that cases are due to the bug of interest; if it looks like a camel, slobbers like a duck and walks like a duck, then it is a MERS-CoV infection yeah? No. If you can clinically characterise and laboratory test then you will more often know the virus the patient has/had than if you don't test. But I'm sure that's clear to everyone anyway.

For MERS, as for H1N1pdm09 influenza and perhaps SARS, finding a reliable pathognomonic set of signs or symptoms capable of reliably distinguishing a respiratory virus of interest from another virus capable of the same disease is not possible. These viruses cause a spectrum of illness. Testing is paramount if you want to know what's there and to address other aspects relevant to public health during an infectious disease cluster/outbreak/pandemic. There are a couple of issues here (at least!)...

From a patient management perspective, who really cares what is making my patient very ill anyway? It really doesn't matter right now if it's this respiratory virus or that one; there are few vaccines and I don't have an antiviral for most of them anyway. I and my healthcare team are already taking respiratory infection precautions and I just want to direct my supportive therapy and resources to the problems they have, right? I'll be (well...you, experienced medical types of which I am not one) doing that before many lab results show up anyway. 

From the perspective of interrupting and understanding viral transmission however, nondescript signs and symptoms are a nightmare. And in the early days of a new virus where we seem to know very little about what path(s) transmission is taking (and perhaps we're also learning some more about those possibilities in general), any infection by whatever method it is empirically determined should, I believe, be recorded as an infection in order to provide the biggest picture possible; a process we have seen unfolding in the United States with its 2 3 detections (1 locally transmitted) of MERS-CoV or its spiky little footprints.

Given that many viruses cannot be distinguished by signs and symptoms alone, a clinical diagnosis to define a case is less reliable than any pathogen-specific laboratory test. I hope the WHO alters their case definition in the near future. Infectious disease is always teaching us - seems we learned a heap from SARS but even the relatively a few cases of MERS are presenting interesting issues and testing us in new ways. 
[While the US antibody-positive result above has since been retracted, I stand by these comments-Ab testing requires rigor, but that can be provided using several assays and applying a good understanding of Ab technologies and limitations to produce reliable results]

Anyhooooo...been stewing on that for a few days apparently. Let's move on and have a look at the 3 updated charts below. 

We are definitely through to the other side of the Jeddah outbreak (see weeklies chart). While cases do keep accruing each and every day (see dailies chart from 20-March), the downward trend of smaller numbers of illness onsets each day also continues. 

Weekly MERS-CoV detections.
Click to enlarge.

Daily MERS-CoV detections from 20-March.
Click to enlarge.

For perspective on the size and the influence of what 1 hospital cluster can turn into and how that can influence how a virus "looks", take a gander at the extent of the April outbreak. Case are still falling out into April as we get more data. If you look at the monthlies chart at the bottom, I've readjusted that y-axis scale again such that it's maximum value is now 10x higher (350 vs 35) than the scale used for 2012 or 2013's charts. May's tally is currently 4x greater than any month from 2012 or 2013. 

What does MERS-CoV hold for us in the coming months? 

Daily detections of MERS-CoV, 2012-current.
Click to enlarge. 

Monthly detection of MERS-CoV 2012-current.
Click to enlarge.