Medical Device Link.

Originally Published MX January/February 2006

COVER STORY

Interview by Steve Halasey

Medical device executives continually strive to identify new technologies with clinical value and untapped market potential. When they succeed, the result can be a set of market-changing products that offer significant benefits to healthcare professionals and their patients. Sometimes those products even constitute a solid foundation for an entirely new company.

That's certainly been the case for SonoSite Inc. (Bothell, WA), the company that pioneered the development of hand-carried ultrasound devices for clinical applications and has taken a leadership role in developing the markets for such devices. SonoSite's systems rely heavily on the advances in application-specific integrated circuits (ASICs) that were initiated during the 1980s and 1990s by its parent company, ATL Ultrasound.

SonoSite president and CEO Kevin M. Goodwin on fast growth and great prospects for hand-carried ultrasound Photo by Scott Levy

Now president and CEO of SonoSite, Kevin M. Goodwin took charge of ATL's hand-carried ultrasound research project in February 1997. A little more than a year later, in April 1998, he presided over the spin-out of the project to form SonoSite, a new company launched as a public entity before it even had a working prototype of its product. Goodwin admits that the move was a risky one. "Although we had a strong hypothesis that the device could be generally useful in medical applications, we had no clue who would want to buy it—if anyone," he says.

In October 1998, SonoSite's engineers unveiled the prototype of the company's first-generation hand-carried ultrasound system, the SonoSite 180, which was launched commercially in September 1999. Since then, SonoSite has advanced the technology base for hand-carried ultrasound to a third generation, embodied in its latest system, the MicroMaxx, which was launched in April 2005.

In this interview with MX editor-in-chief Steve Halasey, Goodwin describes the roots of his company's dominance in the field of hand-carried ultrasound, how the company sets priorities to achieve growth, and the opportunities that lie ahead for expansion into both new clinical applications and new regional markets.

MX: You were with ATL Ultrasound when the company began to develop the technology that now provides the foundation of SonoSite. It's a chicken-and-egg question, but which came first for ATL, the technology breakthrough or the identification of a clinical need?

Kevin M. Goodwin: Technology performance led the move, and the main architect was Jacques Souquet, PhD, who was the chief technology officer of ATL in the late 1980s. Souquet started the whole movement of ultrasound toward digital beam forming, which involves using a transducer to send a sound beam into the body and then processing the microwave forms that bounce back into a single image.

For 10 to 15 years, ultrasound machines accomplished all this complex signal processing using analog circuits. But the result was that the machines got bigger and bigger because the more circuits they contained, the more information they could process and the better the resulting images. So ultrasound got big—and expensive.

What happened to disrupt the growth of ultrasound into even larger and more expensive analog systems?

In the early 1990s, Souquet pioneered ATL's move to displace the older and larger analog circuits by using digital circuitry. The circuitry was etched onto a silicon chip in a binary format—lots of zeros and ones—and then controlled using software. So ATL actually started introducing all-digital cart-based ultrasound systems in the early 1990s.

As techniques for manufacturing integrated circuits improved, enabling more and more circuits to be put deeper and deeper on a silicon chip, Souquet started a skunk works project to develop an ultrasound device that could be carried. By the end of 1995, that skunk works turned into a project funded by the U.S. government's Defense Advanced Research Projects Agency (DARPA), to create a product for the military.

One result of this background is that our very first product—and every one of our products since then—was built to satisfy military specifications (MIL-SPEC) for battlefield use. So, for example, all of the products meet military specs for waterproofing, durability, and reliability. Right now, the U.S. military and other militaries around the world account for several hundred systems.

Was DARPA's primary interest always for the medical applications of this technology?

The DARPA grant came as part of the Department of Defense (DoD) Battlefield 2000 initiative, which was intended to develop ways to reduce battlefield mortality through interventions during the so-called 'golden hour'—the hour immediately after a soldier is wounded. The idea was that if medical personnel could image a patient's thorax and abdomen during that first hour, they would be better able to save the patient's life. So the driving force was to create a device that could go in a medic's backpack and be used during the golden hour, even under battlefield conditions.

But along the way, we have always viewed the technology as potentially useful for a bunch of other industries. The Norwegians and Dutch have considered it for looking at cheese thickness. European Union authorities have studied it for noninvasive imaging of the stomachs of people suspected of trying to smuggle drugs by swallowing them. And there has been interest in the technology for nondestructive fatigue testing, looking at subtle cracks in load-carrying beams and similar components.

How did you become involved with this project?

In December 1996, I had dinner with Dennis Fill, the chairman and CEO of ATL, and he asked me if I would be interested in taking over commercial strategy development for the handheld ultrasound group. I shifted over to the group in February 1997 and began doing research all over the place—internal, external, formal, informal. We just ran all over the world and tested out the notion of a hand-carried ultrasound machine.

What I did not anticipate was some of the challenges of leading the unit as a public company. Most people would probably say that SonoSite should never have been born public. We had a lot of learning to do, and we did it as a public company. That caused some moments of both joy and sorrow. But overall, taking over SonoSite has been a great move for me.

Federal funding usually comes with fairly specific requirements for the achievement of certain milestones. What kinds of expectations were placed on the original project?

There were milestones, as there always are. The most important one was that by October 1998 we had to have an imaging system designed against the assigned specifications. And in addition to solving the problem that the military specified, the company also had to put the technology to work commercially in order to create jobs. We did both.

Growing a New Company

At the end of 1997, SonoSite's technology was just beginning to take shape. But the company went public the following April, even before it had a working prototype. What were the factors that went into making that decision? Was it a difficult call to make?

It wasn't all that difficult, for a couple of reasons that ATL's CEO wisely recognized. First, inside a large, cart-based ultrasound company, a technology project like this would probably be killed off or at the very least held back. Disruptive technologies such as these never see the light of day if there is any risk that they might cannibalize the company's core business.

Second, in order to move the needle forward, we needed to spend more money on the new business. That need was inconsistent with ATL's focus on increasing its profitability.

So there we were: born public. It wasn't by any means a snap decision, but in the long run I think it was a very good decision.

What kind of intellectual property (IP) foundation was provided for SonoSite, and under what terms? For instance, did ATL retain any ongoing shares in the new company?

As part of the spin-out we executed a very specific licensing agreement for market rights to our technology and intellectual property. The market rights were defined in two areas: hand-carried ultrasound, which are devices weighing less than 10 lb, and highly portable ultrasound, which are devices weighing under 15 lb. Within those two defined categories, we have full use of all the pertinent technology and IP that ATL developed through April 2001—that is, three years after the spin-out.

ATL did not keep any equity in the new company, but it has been getting royalty payments. The rate of the initial royalties was set at 3%, dropping down to the current rate of 1.5% after a few years and eventually ending altogether in September 2007.

Beyond the initial capital that ATL provided, what other capital did you have to work with?

Unfortunately, SonoSite went public in 1998, which was a terrible year for small-cap medical device companies. That year, in fact, I remember people saying that investing in a small-cap medical device stock was like going into a roach motel: once you get in, you can't get out. But we raised money in 1999 and in 2001, and we even raised $45 million in the middle of 2002, which was one of the worst stock markets most people had seen in years.

All together, we have raised more than $140 million of capital in the public markets. I wouldn't say it was easy because it was a lot of work, but we did not have much trouble doing it. So today we have $65 million in cash, no debt, a very strong asset base and balance sheet, more than $100 million in annual revenue, 25,000 installed systems, and 470 employees. That's a pretty strong foundation for future growth.

Once SonoSite was independent, how did you prioritize your spending to grow the company and expand adoption of the technology into the variety of fields that you wanted to enter? In what order did you scale up the varied types of personnel that the new company required?

First we had to make some decisions about our market priorities. In 2000, we were selling to probably 25 different markets just because that many markets were interested. But during the course of that year, we whittled down our efforts to focus on just four core markets. In 2001 and 2002 we expanded our reach to six markets, and today we are focusing on just six to eight areas that I would call core markets.

Second, from a technology standpoint, we knew that we would need to turn over our ASIC chips to a new generation every two years, because that would enable us to increase performance while minimizing the size and cost of our devices. Knowing that these technical improvements would enhance the marketability of our products obviously made them a high priority for company spending.

Today we are in 75 markets. Direct sales in nine of those markets, including the United States, contribute about 70% of the company's revenue. In the rest of the markets we sell through dealerships or strategic partnerships.

Balancing Financial Priorities

In prioritizing SonoSite's growth strategy, were there ever any decisions that you felt were particularly risky—ones that could have taken the company down the wrong track or cost a lot of money?

When we launched our first product, in September 1999, we started out by using a distributor to sell our products. We did that for two quarters, and pretty quickly realized that it was a major-league wrong decision. So the good news was that we exited that decision pretty quickly.

But when we started direct selling, in February 2000, we did not know whether we would be able to gain any traction. We hired 28 people, and that year they produced about $8 million in revenue. By the end of that year, we figured out that we needed to focus on a smaller number of markets, and we needed about twice as many reps as we had originally hired. So in 2001 we doubled our sales force and produced $23 million in revenue—a 3X gain. In 2002 we moved up to $45 million in revenue, and we have been stepping up every year since.

SonoSite currently has similar distribution relationships overseas, but it is starting to convert those to direct sales. Is that correct?

Yes, at least in those markets where we think there is enough scalability to make it worthwhile. So we have migrated toward direct selling in Australia, Canada, France, Germany, Japan, Spain, and the UK because that approach works best. In those markets, it is more effective to sell directly, using our own people, and to focus on the particular customers we're most interested in reaching. And we're able to sell at high gross margins.

When did you first begin to suspect that your product was likely to become a commercial success?

Almost from the first moment that we began shipping product. From the very beginning, we have received stories that are just amazing.

At the University of California, Irvine, Medical Center, a 19-year-old kid was brought into the emergency room and presented as dead on arrival. But the attending physician used a SonoSite device to look at his heart real quickly, discovered that there was still some activity, and then determined that he had a tear in his heart. The hospital stabilized him, and the kid walked out of there three days later.

We have received literally hundreds of stories like that. That's one of the reasons that our growth hasn't been linear, because all on their own people keep finding places where this technology can make a valuable contribution.

Do you think the physician office market offers a natural area for growth?

Yes. Wherever it's used, this technology is very beneficial. Having an endocrinologist look at a woman's thyroid on her first visit is very beneficial.

In the breast cancer arena, a woman with a palpable lump gets referred to a surgeon. In the old days, it took as long as 50 days to get a diagnosis. Now, the breast surgeon can look at the lesion with ultrasound and, if need be, perform a biopsy right there in the office. This reduces the wait time by as much as half.

Amazingly, with all the money we spend on healthcare, it is still unusual for doctors to look at their patients' carotid arteries and aortas to make sure they are in good arterial health. But subtle changes in the thickness of the carotid arterial wall are a great predictor of whether a person is going to develop arterial plaque in a few years. In five years, we think it will be a common practice for primary care physicians to perform this kind of testing. And SonoSite expects to drive the adoption of this practice by offering devices that are simple, efficacious, easy to use, and cost-effective.

When it comes to adopting ultrasound for new applications, how responsive are physicians? If you decide to educate a certain specialty area about new possibilities, do you see an increase in uptake there as opposed to other fields?

Yes. That's the thing about this that is so exciting. The more we educate, the more uptake we get. We have concentrated on hospitals. Hospitals have the highest level of physician traffic, and that is where the uptake has been the best. In looking at the clinical markets and considering how many physicians and hospitals are out there, our penetration is still pretty marginal. But it is growing rapidly. We haven't spent as much time focusing on the physician-office market, but we will in the future.

Self Study

Does SonoSite sponsor its own clinical trials, or is it primarily supporting others who are studying its products?

We encourage entrepreneurial and innovative medicine, and we periodically support small, simple studies. An increasing part of our portfolio is sponsoring application-specific trials of various sizes. For example, a New York hospital conducted a study of the use of ultrasound on inbound patients for triage, and it had some very nice outcomes. There was a time-and-motion study done in Germany comparing our device with a cart-based ultrasound unit in terms of saving time. There was a study done in a hospital in the UK in which they used our device to reduce queuing in their public healthcare clinic. All the outcomes have been excellent, very favorable to SonoSite.

The SonoSite Web site offers physicians guidance about how to code for reimbursement. What kinds of regulatory and reimbursement challenges has the company faced?

In 2003, some of the Medicare carriers decided they weren't sure they were going to reimburse hand-carried ultrasound anymore. That was due to a lack of information about our technology's applications and what it could do. The carriers did not realize the true capabilities of our ultrasound systems and the fact that they can provide a complete ultrasound exam. Eventually, we were able to get that problem remedied. In fact, the process of educating those carriers turned out to be a very positive learning experience. It made us realize that reimbursement considerations need to be one of our priorities.

Is SonoSite's current study of physician office assessment of carotid atherosclerosis intended to lead toward completely new clinical applications and common procedural terminology (CPT) coding?

Yes. That is an example of an area where we will indeed one day pursue new coding once the data proves what our hypothesis suggests to be true. That hypothesis is that this application for measuring carotid intima media thickness (IMT) is valuable and will directly affect outcomes. We think IMT measurement could be a standard of care one day in terms of fundamental patient management in cardiovascular disease.

The Next Generation

SonoSite was launched on the basis of some significant intellectual properties (IPs). How has the company developed those and what do you see as the key IP achievements of the company since it was spun out?

We started off with about three patents and six pending. Now we have 35 patents and a number of others are pending. On top of that, we have built a tremendous amount of know-how while building our patent portfolio. I'm not sure which is more important, the legal IP or the actual know-how. But the know-how is what really allows us to go forward and conquer.

SonoSite's big news in the middle of last year was the launch of its third-generation system, the MicroMaxx. How long will it take for the company to attain broad penetration and adoption with that system?

As I've told some of our investors, we are not selling iPod nanos here. It is going to take us a full year, starting in the second quarter of 2005 and running all the way to the second quarter of 2006, to get the technologies fully launched in all the geographic markets in which we are interested.

Does SonoSite's intent to update its chip base every two years or so also set the pace of the company's pipeline for future products?

Yes. Because we are developing an ultrasound product line at much the same rate that personal computers evolve, we usually launch meaningful upgrades every six months and new platforms every two years. For example, we launched the MicroMaxx in the second quarter of 2005. We followed that with a major, meaningful upgrade.

On the Horizon

What percentage of revenues does SonoSite currently invest in R&D?

Historically, it has hovered between 7 and 9%. But we don't really think about it in terms of revenue percentage. We do it on a project basis by examining initiatives we want to pursue. That is what drives our decisions.

The bigger challenge is that we have an abundance of market opportunities, and every year we have to decide which ones we are going to develop and which ones we aren't. For example, we are doing very well in the emergency-medicine market, as well as other acute-care and critical-care markets. But SonoSite has had to spend money up front to stimulate and educate those markets. We have to raise a market's awareness before we can reap the benefits of our investment. There are a lot of markets SonoSite could target, but there are only so many initiatives the company can fund.

The cardiovascular disease management domain is one we think is a very exciting long-term opportunity. By itself, that market could be bigger than the company is right now, but we are pacing our investment in that area as we grow.

Is SonoSite's technology and market lead sufficient to enable it to remain an independent company for the foreseeable future?

Absolutely. We have all the assets of a large company. We have freedom to use as many as 200 Philips ATL patents in the realm of hand-carried ultrasound, which was part of the spin-off agreement. We have very good engineering talent. We have increasingly strong marketing and sales talent, including a large organization completely dedicated to hand-carried ultrasound. And we have a big balance sheet based on a commanding 60% market share in this category.

A couple of the big imaging companies are thinking about how to enter the market for hand-carried ultrasound. But it is a marginal new-product area for them, which they would have to pursue along with many other product areas. Two or three small companies are also thinking about entering the market, but they don't have the capital or the horsepower that we have.

Profitability and Beyond

SonoSite turned the corner into profitability in 2004 and expects to report the same for 2005. Does that achievement make it easier to prioritize spending and reinvestment in the company? Or, now that you really have money, does it make it harder?

It's always hard. It's a matter of probabilities. You have to assign a probability to your projected revenue-based on your pricing and margins-and then you have to decide just where and how much you are going to invest in order to capture that revenue. At this stage, the changes just require much more discipline in how we manage going forward. But the management team is completely committed to being increasingly profitable, so that is the new task we are taking on.

Has profitability caused any change in investor sentiment?

SonoSite has enjoyed a strong cadre of long-term investors for a long time. A few investors that purchased stock in 1998 have remained very strong, and others have come in since that time. So we have a strong base of long-term money in the stock. The company also has the usual amount of short-term trading investors. If things aren't perfect on a quarterly basis those investors tend to trade out, and then they trade in on momentum. And that's just true for all public companies.

Overall, SonoSite's investors are happy with the company's progress. Of course, they want us to deliver profits now along with growth. And that is what we are going to do.

In the future, do you expect SonoSite's overall growth rate to determine the pace at which it releases new technologies, or does the advance of the technology have a pace of its own?

I think the pace will maintain itself because we have a lot of technical assets that have been built cumulatively over time. The first generation of chips led to the second and third generations, and now we have four products out in the marketplace. So we have a lot of knowledge to work from and a lot of base opportunities that we can leverage-in addition to new technologies that we create. Overall, we are in a very good position that way.

Copyright ©2005 MX