A quantitative analysis of technological innovation in minimally invasive surgery

Archie Hughes-Hallett, MRCS1, Erik K Mayer, PhD1, Philip J Pratt, PhD2,

Justin A Vale, MS1, Ara W Darzi, FRS1,2,3

1. Department of Surgery and Cancer, Imperial College London2. The Hamlyn Centre, Institute of Global Health Innovation, Imperial College

London3. Centre for Health Policy, Institute of Global Health Innovation, Imperial

College London

Corresponding author

Erik Mayer,Department of Surgery and Cancer,St Marys Campus,Imperial College London,W2 1NYe.mayer@imperial.ac.uk

Submitted as original research

1

1

2

3

456789

10

11121314151617

18

19

20

21

22

23

24

25

26

27

28

29

30

31

Abstract

Background

In the last 30 years surgical practice has undergone dramatic changed due to the

advent of minimally invasive surgery (MIS). This paper chronologically, and

quantitatively, examines the changing surgical landscape, examining the technologies

that have played, and are forecast to play, the largest part in this shift in surgical

practice.

Methods

Electronic patent and publication databases were searched over the period 1980-2011

for ("minimally invasive" OR laparoscopic OR laparoscopy OR "minimal access" OR

"key hole") AND (surgery OR surgical OR surgeon). The resulting patent codes were

allocated into technology clusters. In addition technology clusters having been

repeatedly referred to in the contemporary surgical literature were also included in

analysis. Growth curves of publications and patents for the resulting technology

clusters were then plotted.

Results

The initial search revealed 27,920 patents and 95,420 publications meeting the search

criteria. The clusters meeting the criteria for in-depth analysis were: instruments,

image guidance, surgical robotics, sutures, SILS and NOTES. When examining the

respective technology clusters, three patterns of growth were observed: a classical S-

shape (instruments and sutures); a gradual exponential rise (image guidance and

surgical robotics); and a rapid contemporaneous exponential rise (SILS and NOTES).

2

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

Conclusion

This paper has revealed technological innovation in MIS has been largely stagnant

since its initial inception nearly 30 years ago, with few novel technologies emerging.

This said, there has been a recent and sustained spike in innovation surrounding SILS

giving weight to the claim that it represents an important part of the future landscape

of MIS.

3

57

58

59

60

61

62

63646566676869707172737475767778798081828384858687888990919293949596979899

Introduction

Healthcare Innovation can be defined as ‘a dynamic and continuous process

involving the introduction of a new technology or technique that initiates a

change in practice’.1–3 In the last three decades, surgical practice has undergone

radical change with the move from conventional to minimally invasive surgery (MIS).

This transition has brought with it a change in the way in which surgeons undertake

their operating practice. This change has been driven, at least in part, by

technological innovation. Since the mid 1990s, this innovation in MIS has been

largely incremental punctuated by disruptive changes in approach, an example of a

disruptive change being the advent of robot assisted laparoscopic surgery.4 More

recently, two novel approaches to minimally invasive surgery have been proposed in

the literature but have not yet been widely adopted into surgical practice: single

incision laparoscopic surgery (SILS) and natural orifice transluminal endoscopic

surgery (NOTES).5–9

Patents are the initial step in the commercialisation of a concept or technology and as

such patent counts represent a good metric with which to measure technological

innovation.3 In addition to being both reliable and relevant measures of innovation,

patents are readily available on publicly accessible databases. As measures of

innovation diffusion,10,11 patent and publication activity have been widely examined in

the social science literature 10–14 but have only recently been applied and validated for

the assessment of healthcare technologies.1

The aim of this analysis was to utilise patent and publication data to address two

broad aims. Firstly, to objectively establish the major areas of technological

4

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

innovation within MIS since 1980 and, secondly, to assess the innovations that have

been postulated, within the surgical literature, as the major emerging technologies in

MIS (robot assisted laparoscopic surgery, SILS and NOTES).

5

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

Methods

The methodology utilised in this paper is based on previously published work

proposing and validating patents and publications as metrics for innovation in

healthcare technology. When correcting for the previously demonstrated, exponential

rise in patents and publications, the formula below was used. The formula below was

used to correct for the exponential increase in publication and patenting over

time.1

II inormalised=

II ioriginal

ci

c i=t i

t 2011

IIi = innovation index i = year in question ti = total number of patents granted by US patent office

ci = innovation constant (modified from Hughes-Hallett et al.1)

Once the corrected year-on-year counts for publications and patents (i.e. the

innovation indices) had been collated, growth curves for each of the respective

technology clusters were plotted. In addition to individual growth curves, composite

charts displaying the patent and publication activity of all the investigated

technologies were generated to illustrate the chronicity of technology development in

MIS.

Establishing top performing technology clusters by patent filings

Initially, a search was performed of the DOCDB (European patent office master

documentation database) patent database15 using the proprietary software package

6

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

PatentInspiration (AULIVE, Ypres, Belgium). A Boolean search strategy specific to

MIS (Appendix 1) was used to establish patenting and publication activity within the

time periods 1980 to 2011, and 2000 to 2011. The result of the patent search was then

used to create lists of the top 30 performing patent codes for the two time periods.

These two time periods were compared to highlight areas of contemporaneous

innovation.

Once generated, these top 30 codes were sorted into related surgical technologies by

two authors (AHH and EKM) with any difference in opinion arbitrated by a third

(JV). Only well defined technology clusters, not pertaining to specific surgical

subspecialties, were selected for in-depth analysis. To identify any patents within

these technology clusters not captured within the top 30 patent codes, a Boolean

search of the DOCDB was undertaken specific to each cluster (Appendix 1). Using

the same search strategies a further search of the PubMed database was also

undertaken to generate a measure of year-on-year publication activity.

In addition to the technologies identified in this step, clusters that have been

repeatedly referred to in the contemporary surgical literature as areas of potential

growth (SILS, NOTES and robot assisted laparoscopic surgery) were added to the

growth analysis.

Data Analysis

Patent and publication data were plotted against each other in order to determine the

nature of their relationship. Depending on whether the relationship was linear or non-

linear Pearson’s (r) or Spearman’s rank (rs) correlation coefficients, respectively, were

7

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

used to determine the degree of correlation between patent and publication numbers.

Data analysis was undertaken in GraphPad Prism (GraphPad Software Inc, CA,

USA).

8

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

Results

Overall trends in patenting and publication

The initial search of patenting and publication databases revealed 27,920 patents and

95,420 publications pertaining to MIS since 1980. When corrected, the growth in both

patents and publications were found to be highly correlated (rs = 0.949) following an

S-Shaped pattern of growth (Figure 1).

Top performing technology clusters

The top 30 performing patent codes for the period 1980 to 2011, are summarised in

Appendix 2. The area in which the greatest number of patent codes has been citied

was, perhaps unsurprisingly, minimally invasive surgical instruments, making up

53.0% of patents falling with the top 30 (Table 1). The other areas fulfilling the

criteria for growth analysis were: sutures, image guidance and surgical robotics

(Appendix 2).

When the search was restricted to the more contemporary time frame of 2000 to 2011

no new technology clusters emerged. Despite this Among the clusters analysed, the

dominance of instrument innovation appeared to have waned somewhat with surgical

robotics and image guidance seeing increases in their patent share amongst the top

performing codes.

Growth in the top performing and literature-derived technology clusters

Corrected patent and publication counts were plotted against time for the six

technology clusters identified within MIS (minimally invasive surgical instruments,

9

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

sutures, surgical robotics, image guidance, NOTES and SILS) in order to establish

their individual growth curves (Figure 2).

Across the six technology clusters, three differing patterns of growth were observed.

Within instruments and sutures, an S-shaped growth curve was observed. For the

instrument cluster this initial sigmoid curve was followed by a period of new

growth. while Surgical robotics and SILS both demonstrated exponential growth,

starting in the early 1990s and 2005 respectively. starting in the mid 1990s and

continuing up until 2011. While the growth curves for SILS and NOTES both

demonstrated contemporary and rapid exponential growth. image guidance and

NOTES both demonstrated a period of exponential growth followed by a drop

off in the number of patents filed.

When assessing the correlation between patent and publication activity within these

groups, instruments, sutures, surgical robotics and image guidance all demonstrated

strong correlation (rs = 0.929, 0.855, 0.937, 0.945 respectively). All were statistically

significant with p < 0.001. NOTES and SILS demonstrated lower correlation with rs =

0.609 (p < 0.001) and 0.532 (p = 0.002) respectively.

The chronicity of minimally invasive surgery

In addition to plotting the growth curves for specific technology clusters, individual

clusters were plotted alongside one another to garner an understanding of the

chronicity of technology development in MIS (Figure 3), four year moving averages

were used to allow for a better understanding of trends. This demonstrated that image

guidance, sutures, instruments and surgical robotics all had exponential phases of

10

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

growth beginning in the late 1980s. Both sutures and instruments reached a plateau in

growth by the mid 1990s, corresponding to the rise in publication and patenting

activity in MIS overall (Figure 1). As previously mentioned, from their take off in the

late 1980s, image guidance and surgical robotics have seen a sustained, albeit

shallower exponential rise in activity. From 1990 until the arrival of NOTES and

SILS in 2005, no rapid take off is seen in any of the technologies examined. In 2005,

both NOTES and SILS see the beginning of a rapid increase in patent and publication

activity. This activity is sustained for SILS, but for NOTES plateaus in 2009.

11

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

Discussion

This paper has chronologically and quantitatively examined innovation in MIS,

scrutinising technologies identified using a previously published methodology1 in

addition to those that have recurred in the recent literature.5–9 Three patterns of growth

(rapid exponential growth followed by a plateau, prolonged exponential growth and

finally rapid contemporary exponential growth) were identified, each of which

contained technologies exhibiting unique characteristics. When examining the

chronicity of technological innovation in MIS it was found to be polarised with the

technologies experiencing rapid exponential growth found at opposite ends of the

time period examined, with a period of innovation stagnation apparent in between

these two poles.

Within the social science literature, the concept of quantitative analysis of innovation

utilising patent and publication-based metrics has been extensively investigated.12,16

However, quantitative research in the medical literature is limited to two papers:

Trajtenberg’s3 paper examining the value of patents as measures of healthcare

innovation and Hughes-Hallett et al.’s recent publication ‘Quantifying innovation in

surgery’.1 These two papers approach the problem of quantifying innovation quite

differently. Trajtenberg’s work examined the value of patent data within a single

specific technology;3 while Hughes-Hallett et al.’s work offered a mechanism with

which to identify and predict emerging technology clusters, in addition to offering a

process allowing quantification of an innovation’s current and potential clinical

impact potential.1

12

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

Within MIS there seem to have been three distinct patterns of growth since its initial

inception in the 1980s. The genesis of MIS is associated with the most visible

innovation spike. In this phase we see a rapid exponential growth in publication and

patenting activity surrounding the development of novel surgical instruments and

consumables (represented by the instrument and suture categories). This spike

represents the development of the basic minimally invasive surgical tools, and

correlates closely with the overall growth curve for MIS. Generally speaking, these

technologies are simple and of low cost, accounting for the rapid growth in patent and

publication activity as industry and surgeons, respectively, design and validate novel

and essential tools. Subsequent to this highly correlated, exponential phase of growth,

these technology clusters remain areas of significant innovation, but reach plateau

reaching the point of diffusion saturation in the mid 1990s, as the laparoscopic

surgeon’s ‘tool-box’ becomes saturated. At this point any new patents or research

tend to pertain to the refinement of existing technology rather than inception of new

devices.13 This point of diffusion saturation is represented by the plateau of a classical

S-shape growth curve (Figure 4).10 In the mid 2000s a new phase of growth is seen

in both MIS overall and within the instrument cluster, this spike probably

pertains to the adoption of new approaches to MIS (robotics, SILS and NOTES)

and the novel instrumentation they require.

Perhaps more interesting were the trends seen in the remaining technology clusters

examined, with robotics and image guidance exhibiting a different pattern of growth.

These technology clusters begin their exponential phases of growth at a similar time

to the previously discussed group but, in contrast to rapid exponential growth they

have experienced a prolonged exponential growth phase, and in fact appear not to

13

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

have yet reached the point of diffusion saturation after more than 15 years. The

reasons for this are almost certainly multifactorial, but two factors in particular are

worth discussion. First, the nature of these two technology clusters means they pose

numerous and complex engineering challenges when compared to the other

clusters of technology examined, perhaps resulting in a slower rate of development.

In addition to this they also represent ‘non-essential’ technologies they serve only to

augment the practice of MIS rather than providing the tools necessary to

undertake it, and as such can be viewed as ‘non-essential’, thereby garnering less

resource from industry than the comparator and more fundamental technologies.

The final growth pattern was one of contemporary rapid exponential growth, and was

seen within the literature-derived technology clusters of NOTES and SILS. These

technologies had a relatively low number of overall patents, and correspondingly

lower correlation coefficients. When examining the growths of the technology

clusters individually it appears that SILS is undergoing a sustained and rapid

exponential growth, implying innovation growth, while the growth of NOTES is

stalling with the number of patents and publications surrounding the technology

beginning to plateau after 2009. This plateau in patent and publication number within

NOTES would suggest a dwindling of innovation and interest in the subject, and may

reflect a failure of NOTES’s to cross ‘the chasm’ that exists between the innovators

and the early adopters (Figure 4). The concept of a diffusion ‘chasm’ was first

proposed by Everett Rogers and represents the point a technology must translate from

a research to normal clinical environment.10 This chasm may also be responsible for

the recent downturn in patenting surrounding image guidance in MIS, which

14

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

despite prolonged and sustained innovation and investment has failed to

translate into widespread operative practice.

When examining the chronicity of technology, an observation has been the apparent

failure of any novel clusters of technological innovation to emerge in the more

contemporary period examined. All of the innovation clusters unveiled by the

systematic search of the patenting database saw the beginning of their growth curves

in the late 1980s and early 1990s (Figure 3) with no new technology clusters being

identified in the period 2000-2011. This failure to identify any new clusters may be

in part down to the failure of the methodology to identify potentially important

innovation in its nascence,1 but equally, and perhaps more likely, this may

suggest a stagnation in innovation with few, if any, novel technologies having had a

significant impact on minimally invasive surgical practice. This would fit with the

hypothesis put forward by Riskin et al. which proposed that enabling technology

shifts such as MIS are rare occurrences with the remainder of innovation being

incremental in nature.4 Another potential explanation for this stagnation is increasing

medical device regulation. The regulatory process for novel devices is, both in Europe

and the United States, significantly more arduous than for those that are similar to pre

existing technologies. This approach to medical device regulation has the potential to

stifle innovation with device manufacturers likely to shy away from novel

technologies due to the increased financial risk associated with the regulatory process.

Although the methodology proposed here offers a quantitative approach to defining

past, and assisting in assessing future technologies of influence in MIS, it is not

without its limitations. First amongst these is the surrogacy of the measures used.

15

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

To truly establish the diffusion, and as such the success, of a given innovation or

cluster of innovation an assessment of the proportion of patients in which that

innovation has been used must be measured. Although this represents the gold

standard approach when looking macroscopically at innovation within MIS it is

impractical due to the huge number of innovations to be examined. In addition,

the way in which the data is constrained by the search terms used, relying on terms

being both specific and sensitive enough to generate meaningful results. In this

respect the methodology is imperfect in a similar fashion to that of systematic

reviews, which are an accepted and valued part of evidence synthesis.

Looking to the future of MIS the data presented herein adds objective data to

the previous subjective claims that SILS, a surgical technique that has as yet

only been adopted by very few, represents a significant part of the future of MIS.

This transition from specialist to mainstream practice will most likely be

facilitated by improvements in the tools used to perform the technique, with

robotic assistance perhaps the most likely to provide this technological segway.

Conclusions

This paper has undertaken a quantitative and chronological assessment of the patent

and literature databases. Analysis of these data has allowed an accurate and

chronological view of innovation in MIS, revealing it to have been largely

incremental since the advent of laparoscopy with few novel technology clusters

emerging in the last decade. Looking to the future of MIS the data presented herein

adds weight, and objective data, to the previously subjective claim that SILS

represents a significant part of the future of MIS.

16

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

References

1 Hughes-Hallett A, Mayer EK, Marcus HJ, Cundy TP, Pratt PJ, Parston G, et al.

Quantifying Innovation in Surgery. Ann Surg. 2014 Aug; 260: 205–211.

2 Rogers W, Lotz M, Hutchison K, Pourmoslemi A, Eyers A. Identifying

Surgical Innovation: A Qualitative Study of Surgeons’ Views. Ann Surg. 2013

Jun 19;

3 Trajtenberg M. A penny for your quotes : patent citations and the value of

innovations. RAND J Econ. 1990; 21: 172–187.

4 Riskin DJ, Longaker MT, Gertner M, Krummel TM. Innovation in surgery: a

historical perspective. Ann Surg. 2006 Nov; 244: 686–693.

5 Rivas H, Díaz-Calderón D. Present and future advanced laparoscopic surgery.

Asian J Endosc Surg. 2013 May; 6: 59–67.

6 Aurora A, Ponsky J. Future Perspectives on Scarless Surgery: Where We Have

Been and Where We Are Going. In: Desai M, Gill I, editors. Scar-Less Surg.

1st ed. London: Springer; 2013. p. 341–351.

7 Hafron J, Kaouk JH. Technical advances in urological laparoscopic surgery.

Expert Rev Med Devices. 2008 Mar; 5: 145–151.

8 Ficarra V, Ploumidis A, Lumen N. The infancy of robotic laparoendoscopic

single-site renal surgery: waiting for needed technological improvements. Eur

Urol. 2013 Feb; 63: 281–282.

17

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

9 Lee W-J, Chan C-P, Wang B-Y. Recent advances in laparoscopic surgery.

Asian J Endosc Surg. 2013 Feb; 6: 1–8.

10 Rogers E. Diffusion of Innovations. 5th ed. New York: Free Press; 1962.

11 Ryan B, Gross N. The diffusion of hybrid seed corn in two Iowa communities.

Rural Sociol. 1943; 8: 15–24.

12 Daim TU, Rueda G, Martin H, Gerdsri P. Forecasting emerging technologies:

Use of bibliometrics and patent analysis. Technol Forecast Soc Chang. 2006

Oct; 73: 981–1012.

13 Bengisu M, Nekhili R. Forecasting emerging technologies with the aid of

science and technology databases. Technol Forecast Soc Chang. 2006 Sep; 73:

835–844.

14 Nelson AJ. Measuring knowledge spillovers: What patents, licenses and

publications reveal about innovation diffusion. Res Policy. 2009 Jul; 38: 994–

1005.

15 EPO patent information resource DOCDB [Internet]. Eur. Pat. Off. 2008 [cited

2013 Sep 23]. Available from:

http://www.epo.org/searching/subscription/raw/product-14-7.html

16 Hagedoorn J, Cloodt M. Measuring innovative performance: is there an

advantage in using multiple indicators? Res Policy. 2003 Sep; 32: 1365–1379.

18

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

1980 - 2011 2000 - 2011Technology cluster Patents % Patents %

MIS instruments* 498 53 157 45.1Sutures 65 6.9 11 3.2Image guidance 57 6.1 26 7.5Surgical Robotics 60 3.9 26 7.5Not included in analysis 287 30.6 128 36.8

Table 1. Top performing technology clusters *including laparoscopic ports and trocars

19

464465

466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506

Figure Legends

Figure 1. Patent and publication growth in MIS 1980 to 2011

Figure 2. Growth curves for chosen technology clusters. Clusters in the first row demonstrate a classical S-shaped growth curve, those in the second row have shown a gradual but exponential pattern of growth while the expert identified areas of growth demonstrate steep and contemporary exponential growthWhere two y-axes are displayed the left pertains to the publication innovation index and the right the patent index.

Figure 3. Growth curves for publications and patents displayed to highlight the chronology of technological innovation in laparoscopic surgery. Values displayed are 4-year moving averages. Innovation indices for all technologies are plotted on the left y-axes with exception of instruments on the patent chart, and robotics and sutures on the publication chart, which are plotted on the right y-axes.

Figure 4. The S-shaped diffusion curve in this figure demonstrates the 3 phases of growth in any technological innovation (incubation, exponential growth and diffusion saturation) and matches them to the characteristics of the individual members of the adopting population.10

20

507508509510511512513514515516517518519520521522523524525526527528529530531