- Email: [email protected]
A Simple Mathematical Model for Wound Closure Evaluation
Journal of the American College of Clinical Wound Specialists (2016) -, -–-
1 2 3 4 5 6 7 8 Q1 9 10 11 12 Q4 13 14 15 16 17 Q3 18 19 20 21 22 23 24 25 26 27 28 29 30 31 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 57 58 59 60 61
RESEARCH PAPER
A Simple Mathematical Model for Wound Closure Evaluation nb, Israel Giacamana, Alejandra Vidal Villaa, Hugo Mendieta Zero pezb, Laura E. Meza Trillod, Marıa del Socorro Camarillo Romeroc, Sandra Parra Lo David A. Perez P ereze, Miguel Conchaa, Cesar Torres-Gallegosf, Sandra L. Orellanaf, n-Ampuerog, Ignacio Moreno-Villosladag,* Felipe Oyarzu a
Instituto de Anatomıa, Histologıa y Patologıa, Facultad de Medicina, Universidad Austral de Chile, Valdivia, Chile Facultad de Medicina, Universidad Aut onoma del Estado de Mexico, Toluca, Estado de Mexico, Mexico c Centro de Investigaci on en Ciencias Medicas (CICMED), Universidad Autonoma del Estado de Mexico, Toluca, Estado de Mexico, Mexico d Facultad de Ciencias de la Salud, Universidad de Anahuac, Huixquilucan, Estado de Mexico, Mexico e Universidad del Valle de Toluca, Toluca, Estado de Mexico, Mexico f Instituto de Ciencias Quımicas, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile; and g Departamento de Ciencias & Tecnologıas Farmaceuticas, Universidad de Chile, Santiago, Chile b
KEYWORDS: Diabetic foot; Foot ulcers; Mathematical model; Wounds
Abstract The incidence of ulcers associated to type 2 diabetes mellitus (T2DM) increases every year. We introduce and explore a new mathematical algorithm to evaluate wound-healing in foot ulcers associated to T2DM. Fifteen patients (nine women and six men), mean age of 70 6 16 years were included. The evolution of their wounds followed-up for a period of 18–45 days. According to the Wagner grading system the ulcers were grade I (5 patients), grade II (9 patients), and grade III (1 patient). Clinically, the type of the ulcers was neuroischemic (12 patients) and neuropathic (3 patients). A new parameter is introduced, the ‘‘continuous linear healing rate’’ Dc that was more accurate with higher values and requires less quantifications than usual formulas to make a wound-healing projection. 2016 Elsevier Inc. All rights reserved.
Introduction Wound-healing is a complex biological process that involves the expression of growth factors that promote Conflict of interest/financial disclosure statement: Authors have nothing to declare. * Corresponding author. Instituto de Ciencias Quımicas, Facultad de Ciencias, Universidad Austral de Chile, Isla Teja, Casilla 567, Valdivia, Chile. Fax: 156 63 2293520. E-mail address: [email protected]
various cellular processes, production of new connective tissue matrix, and collagen deposition.1 Studies have shown that it is possible, under appropriate treatments, to achieve wound closure in about 45 days. Decision-making by nurses and physicians during wound-healing may be difficult due to lack of quantitative criteria which allow an accurate diagnosis of the quality of the wound progression. With the presence of a pool of protocols and products possessing the potential to improve wound progression to closure, the selection of one of these may become arbitrary. In order to perform quantitative
2213-5103/$ - see front matter 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jccw.2016.07.002
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
2 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195
Journal of the American College of Clinical Wound Specialists, Vol -, No -
analyses, data acquisition must be carried out, which adds another difficulty to the work of the clinical staff. Data acquisition should be simple and fast to be easily incorporated into healing protocols by nurses. In addition, data treatment and interpretation must not involve complicated protocols; either demand high qualification in disciplines other than medical, such as mathematics, physics, or computation. Stochastic models2 that consider delayed exponential behavior,3 or photogrammetric software,4 furnish reliable methods on the evaluation of diabetic foot ulcers, but the protocols, in terms of algorithms and the software required, may result sufficiently complicated to minimize their use by nurses in hospitals and family health centers, where treatments are normally performed. Evidence showed that monitoring healing progression for 2 or 4 weeks may serve to predict eventual wound closure under appropriate mathematical models.5 Mathematical modeling may also aid in identifying potential methods for disease control and management if wound closure results can be correlated with the wound and the patient’s condition.6 Few mathematical models in the literature deal with the complications associated with diabetes due to their complexity.6 The literature reveals great diversity in the procedures employed to compare wound-healing rates, including variables such as periods of study, type, size and shape of the analyzed lesions, closing rate definition, calculation methods, mechanisms, etc.3,7–13 However, controversy exists in the selection of the adequate indicative parameter of the wound closure rate.8,9,14,15 Assessment of wound status in terms of time is often based on measurements of wound area and, to a lesser extent, of the wound perimeter8,14–17 and the linear healing rate (D) has been often utilized. The D parameter is obtained through a method of calculation based on differences; thus, accurate calculation requires a large number of data to achieve statistical significance. However, the concept of linear healing involves a constant, perpendicular advance of the wound margin toward the wound’s open area, and this feature may be used to propose a mathematical development based on differentials. The aim of this study is to introduce and test a mathematical algorithm for calculation of wound-healing rate from few clinical data obtained by monitoring wound closure in patients with Type 2 Diabetes Mellitus (T2DM).
Materials and Methods The research study protocol was approved by the Ethics Committee of the Medical Sciences Research Center (CICMED), Autonomous University of the State of Mexico (UAEMex). The investigation conforms to the principles in the Declaration of Helsinki, and the patients were fully informed regarding the aim of the study and signed written informed consent.
This was a bi-center, open-label, phase I, descriptive study. A total of 15 consecutive patients with T2DM and foot ulcer up to grade III on the Wagner scale were enrolled in this study. Patients were recruited at the Healing Ulcer Service, Clınica 251, Instituto Mexicano del Seguro Social (IMSS), Metepec, Estado de Mexico (3 patients), and at the Clınica de Consulta Externa ‘‘Alfredo del Mazo Velez,’’ ISSEMyM, Toluca, Estado de Mexico (12 patients) between September 2012 and August 2013. Other subjects who had clinical signs of severe infection and those unwilling to participate in the study were excluded.
Clinical Follow-up Age, gender, duration of diabetes, time of wound evolution, Wagner ulcer grade, and vascular status were recorded for all patients. The ulcer was delineated for each patient on a sterile acetate sheet, and a scale was added. Images were digitized and analyzed with AutoCAD 2009 software to determine their area and perimeter. Five or more measurements were carried out at the beginning of the study and during follow-up for 18–45 days. Patients were advised to have their dressing changed daily, either at the outpatient clinic of our hospital, at a nearby local hospital, or with the aid of the community nurse who provided home care for patients with ambulatory problems.
Mathematical Model According to Gilman,8 we can define the linear healing rate (D) as follows: D52
DA Pavg Dt
ð1Þ
where DA is the change in superficial wound area between 2 consecutive clinic visits, Dt is the time between the two visits, and Pavg is the average wound perimeter measured at the two clinic visits. This equation is evaluated in discrete times during the wound-healing process, normally once a week, and the final value is averaged among the discrete values obtained at each discrete time, obtaining the average D value (Davg). The drawback of this method is that in order to obtain statistical significance, a large number of visits and data acquisitions are needed, and data frequently exhibit great dispersion. However, the concept of linear healing involves a regular advance of the wound contour toward the wound open area and perpendicular to the wound margin; thus, data corresponding to this magnitude are theoretically linked with each other by a linear trend over time. This concept may be mathematically formulated by considering the effective advance of the wound margin during an infinitely small period of time (Fig. 1). Thus, the continuous linear healing rate (Dc) can be defined as:
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262
Villa et al 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
---
Dc 5 lim D5 lim 2 Dt/0
Dt/0
3 DA dA dr 52 52 Pavg Dt Pdt dt
ð2Þ
where A: area, d: delta, lim: limit, P: perimeter, Pavg: average wound perimeter measured at two clinic visits, r: length perpendicular to the wound contour, DA: change in superficial wound area between 2 consecutive clinic visits, Dt: time between two visits. dA 5 lim DA when DA approaches 0. Considering that according to Fig. 1: ð3Þ
dA5Pdr
where A is superficial wound area, r is the length perpendicular to the wound contour, and P is the wound perimeter. Then; it is clear that dA52PDc dt
ð4Þ
Integration of equation (4) leads to A 52Dc t1q P
P2 A
tclosure 5
q Dc
ð7Þ
If we consider linear healing as a physiological fact, deviation from the linear trend following equation (5) should be caused by changes in the shape of the wound and other physiological conditions that may influence the value of Dc during wound-healing.
Statistical Analysis ð5Þ
where q is the intercept, denoting the extrapolated A/P initial values. Plotting A/P vs. time, the values of Dc and q can be obtained from the slope and the ordinate at the origin, respectively. Basically, q contains information on the aspect of the wound, such as those comprising extrapolated A/P initial values. Because P and A have dimensions of length and square length, respectively, we can define the Shape coefficient (Sc) as: Sc 5
This coefficient does not depend on the size of the analyzed figure, but on its shape, and may be an appropriate parameter for evaluating changes in the shape of a particular wound or differences among the shapes of different wounds. This assumes a value of 12.56 for circular wounds and 16 for square wounds, as depicted in Fig. 1. In addition, equation (5) can be used to obtain projected time of wound closure (tclosure), a condition under which A 5 0, so that:
ð6Þ
STATISTICA version 8.0 software was utilized to analyze statistical correlations between variables; statistical differences by grouping patients by Wagner’s grade of wound type were searched using the Mann–Whitney U test.
Results Fifteen patients (nine females and six males) with a mean age of 70 6 16 years were included in the study. According to the Wagner grading system, ulcers were grade I (5 patients), grade II (9 patients), and grade III (1 patient).
Figure 1 Wound evolution scheme for different shapes. A: area, Dc: continuous linear healing rate, lim: limit, P: perimeter, Pavg: average wound perimeter measured at two clinic visits, r: length perpendicular to the wound contour, Sc: shape coefficient, DA: change in superficial wound area between 2 consecutive clinic visits, Dt: time between two visits.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
Q2
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396
Journal of the American College of Clinical Wound Specialists, Vol -, No -
4 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
Table 1
Anthropometric and clinical characteristics of the patients.
Case
Gender
Age
Duration of diabetes (years)
Time with the ulcer (months)
Wagner stage
Wound type
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
M F F F M M F M F F M F M F F
56 62 79 91 85 71 40 74 97 45 75 62 67 66 81
25 4 5 26 35 6 13 14 18 1 5 1.6 5 2 8
12 2 2 3 2 6 63 4 5 6 8 2 3 1 5
II II II II II III II I I II II I I I II
Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuropathic Neuroischemic Neuroischemic Neuroischemic Neuropathic Neuroischemic Neuropathic Neuroischemic
F: female, M: male, P: patient.
Clinically, the ulcer type was neuroischemic (12 patients) and neuropathic (3 patients). Duration of ulcer evolution prior to treatment ranged between 1 and 63 months (Table 1).
Ulcer Healing Data obtained from planimetric measurements have been treated introducing A, P, and t in equations (1), (5) and (6). D values obtained from equation (1) for the healing process of the 15 patients are shown in Fig. 2, as they are usually presented in the literature.15 The corresponding Davg values are listed in Table 2, together with the corresponding values of the absolute and relative Standard Deviations (SD and RSD, respectively). D values of the same order of magnitude as those
Figure 2
usually reported were found.17 However, in the majority of cases, corresponding RSD were .100% (Table 2), with an outstanding value close to 800% for patient (P) 14. RSD was used to evaluate the quality of the data for the Davg calculation, and the high values found indicate low quality. In addition, there was no correlation between the Davg value and RSD values (r 5 0.4834; p 5 0.68); therefore, the data dispersion for each patient was assumed to be independent from the healing rate. High dispersion of D values for each single patient makes it difficult to establish comparisons between patients. On the other hand, utilizing the same raw data, Dc values have been obtained by means of equation (5) and are illustrated in Table 3. In order to obtain trends deriving from the same origin, A/P data and the corresponding linear adjustments have been plotted in Fig. 3 for all patients after
D (?) for each patient. D: linear healing rate.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530
Villa et al 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597
Table 2
---
5
Davg and corresponding SD and RSD.
Patient
Davg 6 SD (cm/day)
RSD (%)
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
0.00828 6 0.01424 20.00701 6 0.03318 0.00939 6 0.01858 20.00868 6 0.02981 0.00916 6 0.00871 0.01094 6 0.00959 0.00263 6 0.00453 0.00740 6 0.01024 0.00580 6 0.01060 0.00795 6 0.01657 0.00244 6 0.00656 0.01012 6 0.00737 0.00484 6 0.00623 0.00153 6 0.01222 0.00401 6 0.00535
172 473 198 344 95 88 172 138 183 208 269 73 129 796 133
Davg: average linear healing value, RSD: relative standard deviation, SD: standard deviation.
subtracting q. The resulting straight lines were distributed in a cone. Expressions of linear adjustments of A/P vs. t and the corresponding R2 values are also presented in Table 3. R2 values were also used to evaluate data quality for calculation of Dc, based on the assumption of a linear behavior. R2 values found presented values between 0 and 1. There was a positive correlation between Dc and R2 (r 5 0.6384; p 5 0.010). Because this correlation was positive and statistically significant, we can conclude that higher Dc values resulted more accurate than lower values. One patient (P2) exhibited negative values of Dc, indicating that the wound was enlarging, and a very low value of R2
Table 3
was found (0.03). Six patients showed positive Dc values of ,0.002 cm/day (P11, P9, P15, P14, P7, and P13), and the corresponding R2 ranged between 0.22 and 0.60. Six patients demonstrated values of around 0.004 cm/day (P8, P12, P4, P1, P3, and P6), and the corresponding R2 ranged between 0.24 and 0.96. Two patients showed Dc values of .0.006 cm/day (P10 and P5), and the corresponding R2 were 0.62 and 0.95. Total wound closure should occur at tclosure (extracted from equation (7)) whose values are shown in Table 3 for patients showing Dc values. The values obtained for tclosure ranged between 52 and 805 days. The mean tclosure value obtained for the patients was 260 days, and the data showed an SD of 212 (RSD, 81%). In order to compare the results obtained from two different methods, Dc and Davg values were plotted in Fig. 4a, where patient entries were ordered according to Dc values. In that Dc and Davg were calculated from the same data, a positive correlation was expected between both magnitudes, which was statistically confirmed (r 5 0.5490; p 5 0.034). However, outstanding differences were found in P2 and P4. For P4, an inversion on the sign of the corresponding value was observed, because Dc was positive and Davg was negative. In general, positive values of both magnitudes were found, with Dc lower than Davg. In addition, there was no correlation between the corresponding R2 (corresponding to the calculation of Dc by linear adjustment) and RSD corresponding to the calculation of Davg (r 5 0.2267; p 5 0.417) (Tables 2 and 3). Notably, for patients exhibiting high values of R2 when calculating Dc (.0.90), such as P1, P5, and P12, the RSD values found were still high and close to 100%. Assays for observing any correlation between Dc or Davg with wound shape and size have been conducted. In this
Linear adjustments of A/P data versus t, Dc, Sc, and corresponding SD and RSD.
Patient
Linear adjustment (equation (6))
R2
Dc (cm/day)
Sc 6 SD
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
y y y y y y y y y y y y y y y
0.96 0.03 0.52 0.24 0.95 0.83 0.60 0.85 0.33 0.62 0.22 0.94 0.82 0.53 0.48
0.00396 20.00080 0.00438 0.00380 0.00686 0.00496 0.00152 0.00351 0.00069 0.00660 0.00034 0.00365 0.00189 0.00010 0.00095
13.74 14.86 13.40 18.54 18.65 14.54 20.23 14.04 15.43 15.64 17.01 15.90 15.10 14.35 14.60
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
20.00396x 1 1.25068 0.00080x 1 0.38320 20.00438x 1 0.65355 20.0038x 1 0.40481 20.00686x 1 0.35369 20.00496x 1 0.68489 20.00152x 1 0.24996 20.00351x 1 0.44695 20.00069x 1 0.22981 20.00660x 1 1.12439 20.00034x 1 0.21049 20.00365x 1 0.53096 20.00189x 1 0.36564 20.00010x 1 0.32471 20.00095x 1 0.76603
6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
0.34 0.89 0.34 0.77 5.35 0.22 2.48 0.31 0.94 0.84 0.48 0.73 0.52 0.22 0.40
RSD (%)
tclosure (days)
2.47 5.96 2.53 4.17 28.7 1.49 12.3 2.21 6.08 5.38 2.80 4.61 3.45 1.52 2.71
316 – 149 106 52 138 165 127 331 170 622 145 193 326 805
Dc: linear healing value, RSD: relative standard deviation, Sc: shape coefficient, SD: standard deviation, tclosure: time of wound closure, y 5 A/P (cm), x 5 t (days), R2: linear regression factor.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664
665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731
Journal of the American College of Clinical Wound Specialists, Vol -, No -
print & web 4C=FPO
6
Figure 3 (A/P) – q as a function of time. Data points and linear arrangements for the 15 patients. A: area, P: perimeter, q: intercept of the formula A/P 5 2Dct 1 q, where Dc: continuous linear healing rate and t: time (days).
regard, four magnitudes related with initial wound shapes and sizes have been evaluated as follows: initial Area (Ainit); initial Perimeter (Pinit ); initial Shape coefficient (Scinit) init init defined as Sc 5 (P )2/Ainit, and q. The respective values were plotted in Fig. 4b. Ainit ranged between 0.78 and 21.1 cm2, and no correlation with either Dc (r 5 0.5015; p 5 0.057) or Davg (r 5 0.2311; p 5 0.407) was found. Pinit ranged from 3.4 to 17 cm, and respective values correlated with Dc (r 5 r 20.5376; p 5 0.039), but did not correlate with Davg (5 0.1687; p 5 0.548). A negative correlation was found between Scinit and Davg (r 5 20.6101; p 5 0.016), but not between Scinit and Dc (r 5 20.3361; p 5 0.221). q values ranged between 0.21 and 1.25 cm, demonstrating no correlation with Dc (r 5 20.353; p 5 0.197) or Davg (r 5 20.272; p 5 0.326). Thus, taking these results as a whole, we can conclude that there is a weak correlation between the healing rates found by both mathematical methods and the initial wound shape and size. Wound shape evolution during this study has been documented by the Sc calculation through equation (6). As previously mentioned, Sc is a parameter that does not depend on the size of the analyzed figure, but on its shape. Likewise, a perfect circle has a Sc of 4p, the lowest possible value to be found (Fig. 1). Sc mean values, together with the corresponding SD and RSD, are plotted in Fig. 5a. In this study, Sc values ranged between 13.1 and 18.2, indicating that wounds were not elongated toward any specific direction. There was no correlation (r 5 0.227; p 5 0.417) between R2 values of the linear adjustments (Table 3) from which Dc was calculated and RSD obtained for each patient during wound evolution, indicating that a change in wound shape may not be a cause of loss of linearity. Similarly, no correlation was found between the RSD values from which
Davg was calculated and RSD (r 5 20.231; p 5 0.408), indicating that data dispersion for any patient was not necessarily caused by a change in wound shape during the healing process. In fact, small changes in wound shape have been registered, a witnessed by low values of RSD, .6%, with the exception of P5 and P7, for which the corresponding RSD yielded values of 28.7 and 12.3%. Evolution in time of this parameter for these two patients can be observed in Fig. 5b. The wound shape for P5 clearly changed after the first 15 days, adopting a more elongated form during the last 2 weeks (Fig. 5c). Evolution of the wound of P7 appears more erratic, demonstrating Sc values trending up and down, as can be noted in Fig. 5b and c. However, evolution of the wound of P5 exhibited excellent linear behavior in the wound-healing process, with an R2 value of 0.95 for linear adjustment of data, whereas an R2 of only 0.60 was obtained for P7. In addition, data for P5 are less dispersed than those for P7 when calculating Davg, demonstrating RSD of 95 and 172%, respectively (Table 1). We also noted that the patient’s health condition reported in Table 1 could not be correlated with Dc or Davg. Furthermore, no statistical differences were found considering Wagner grade (Table 1) and Dc (U 5 17; p 5 0.463) or Davg (U 5 21; p 5 0.841), employing the Mann–Whitney U test. For this test, the sample corresponding to P6, the only Wagner III patient, has not been considered. Similarly, no differences were found regarding wound type (Table 1) and Dc (U 5 16; p 5 0.7728) or Davg (U 5 16; p 5 0.7728), utilizing the same statistical test. In fact, only two of the three neuropathic patients showed Dc values . 0.004 cm/day (P8 and P12). The only Wagner III patient (P6) presented a Dc value of 0.00496 cm/day. Wagner I ulcers corresponded to the three neuropathic patients (P8, P12, and P14) plus P9 and P13, and all patients presented Dc values , 0.004 cm/day. The two younger patients (P7 and P10, aged 40 and 45 years, respectively), showed neuroischemic ulcers, with similar Dc values of around 0.006, despite the duration of both of their wounds and that the pathologies differed from each other.
Discussion The integrated of data acquisition technology for the measurement of wound extension, such as photographic systems with adequate software to record and analyze images, and systems that employ stereophotometry, may furnish additional advantages, such as low interobserver variability. In addition, sophisticated systems are able to distinguish by color analysis areas with granulation and necrotic tissue.18,19 After image treatment, parameters such as A and P may be obtained at different time values. In classical, simple analyses, affordable by nurses and physicians at hospitals and family health centers, healing rates are obtained by calculation of absolute and relative surface
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865
---
7
print & web 4C=FPO
Villa et al
Figure 4 a) Dc (-) and Davg ( ) for each patient; b) Ainit (-), Scinit (,), q values (-), and Pinit ( ) for each patient. Ainit: initial area, Davg: average linear healing value, Dc: continuous linear healing rate, Pinit: initial perimeter, q: intercept of the formula A/P 5 2Dct 1 q, where Dc: continuous linear healing rate and t: time (days), Scinit: initial shape coefficient.
changes in time and calculation of parameter D.17 However, these analyses may fail due to large variations in D values obtained between two consecutive clinic visits. Thus, as what has been shown in this study, the high data dispersion achieved when calculating Davg would not allow for extracting useful information for decision-making with respect to changing treatment, for example, or for applying any specific product. In our approach, considering the ease of data acquisition, planimetry of the wounds on acetate sheets represents a simple and short time-saving strategy (Fig. 6).20,21 The alternative data treatment developed here permits the extraction of hidden information, based on the assumption implicit in the definition of linear healing, i.e., a regular advance of the wound contour toward the wound open area and perpendicular to the wound margin, so that data corresponding to this magnitude are theoretically linked with each other by a linear trend over time. Thus,
the slope of the linear trend, which we denominate Dc, arises from the introduced mathematical model to characterize wound-healing. In addition, the Dc parameter possesses an advantage over Davg that allows for the use of a lower number of values for the analyses, in that the data are linked by a mathematical formula. Also, it provides better visual information because the parameter is represented by the slopes of straight lines in a graph where the wound evolution for different patients can be compared. In order to better understand these terms, the data obtained for P5 are presented in Fig. 7. In treating data according to the classical method, therefore calculating Davg by means of equation (1), the differences in A and P between two consecutive visits, independent from differences between two other consecutive visits, must be considered. Moreover, individual D values are classically calculated only by treating the data, but
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932
Journal of the American College of Clinical Wound Specialists, Vol -, No -
8
Figure 5 a) Mean Sc ( ) (with SD) and respective RSD (-) for each patient; b) Sc values for P5 (?) and P7 (-); c) Wound margins for P5 and P7 (out of scale). P (number): number of the patient, RSD: relative standard deviation, Sc: shape coefficient, SD: standard deviation.
not graphically. We represent the A/P data vs. t in Fig. 7 (top left and right). The individual D values obtained between each pair consecutive visits are represented by the
slopes of the small segments (Fig. 7, top left). The lines do not completely adjust to the segments linking two consecutive data, in that equation (1) employs Pavg instead
print & web 4C=FPO
933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999
Figure 6
A) Photograph and B) its scanned image of one case of foot ulcer.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066
Villa et al 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133
---
9
Figure 7 Top left: Data points obtained for P5 (?), and its treatment according to equation (1); individual D values are graphically shown as the slope of the lines. Bottom left: individual D values are plotted (C), together with the corresponding Davg (-) and its RSD. Top right: Data points obtained for P5 (?), and their linear adjustment according to equation (5). Bottom right: the Dc value and the corresponding R2 obtained are shown. A: area, D: linear healing rate, Davg: average linear healing value, P: perimeter, R2: linear regression factor, RSD: relative standard deviation.
of P. Then, a large dispersion in D values is found, since the trending up and down of the data produces large differences in the slope of the corresponding segments; thus, RSD values are also found to be large (Fig. 7, top and bottom left). Expression (5) has the advantage over expression (1) in that fewer data points are needed to trace the healing profile, because of the theoretical linear trend; therefore, linkage between data, is considered. Thus, from the mathematical point of view, five visits taken once a week where data are acquired should be sufficient to find the linear trend of the wound-healing by fitting five experimental points. The innovative method presented reflects that the data are not independent one from the other, but are linked by a linear trend. Thus, Dc can be easily calculated, with few data, by obtaining the slope of this linear trend after linear adjustment, with R2 as the parameter for evaluating the strength of the linear association among the variables. Note that, for P5, the data followed a linear trend with an excellent R2 of 0.95, which confirmed the good quality of the Dc estimation (Fig. 7, right), whereas an RSD of 95% was obtained for the Davg calculation.
Among information obtained by introducing data on the mathematical algorithm, is the tclosure, a parameter that could be easily taken into consideration to make a clinical decision. For example, it has been reported, for diabetic foot ulcers that, linearity of the healing process may be lost for ulcers healing beyond 28 days.17,22 In this regard, tclosure should be interpreted as an extrapolation of linear healing during the first 28–35 days of healing, and used for a decision to change treatment rather than as a predictive value. In addition, the correlation found between the values of Dc and R2 may supply another criterion for decision-making. Low Dc values with low R2 may indicate poor quality of wound-healing progression. On the contrary, high Dc values with high R2 may indicate that the treatment is effective. For example, if Dc values . 0.003 cm/day and R2 .0.8 are imposed as quality criteria (to this point, arbitrarily), 5 patients (P1, P5, P6, P8, and P12) comply with this requisite. Moreover, these parameters offer means of comparison of the wound evolution of a specific patient before and after changing the treatment, once the latter is decided upon by the medical personnel.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
10 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267
Journal of the American College of Clinical Wound Specialists, Vol -, No -
In clinical setting, mathematical formulas based on the argument that the dimension under assessment conforms to a linear phenomenon23 have been used not only to evaluate wound-healing24 but also scar evolution.25 The data obtained from this small sample of patients suggest that we can predict the type of response to the healing strategy after 4–6 weeks of follow-up. If this is corroborated with a larger study, the possibilities to replicate the prediction of this new formula in other diseases are high, more if we add the use of complex image studies like axial computed tomography (ACT) or Magnetic Resonance Image (MRI). Data accumulation will aid in establishing quantitative criteria for making a diagnosis of the quality of wound progression and for obtaining more accurate parameters and criteria for comparison of the wound-healing evolution for different patients. The incorporation in the future of more technological tools for image capture, processing, and analysis in clinical practice may be slow, but probably unavoidable. It has been reported in the literature that, providing standard care, the time course of wound-healing in diabetic foot ulcers is predominantly determined by etiologic factors, and to a lesser degree by wound size.26 However, both the patient’s health condition and the wound shape have been invoked to explain differences in healing rates.14,15,17 In this study, null or weak correlations have been found between Dc or Davg in terms of wound size and shape, parameterized by means of A, P, and Sc, or with the patient’s condition. A clear limitation of the study is the low number of patients. In spite of this drawback, this was a pilot study and the advantages provided by the calculation of Dc and related parameters may represent an important feature for the medical management of several wound types, because accurate results may be easily obtained solely by introducing a few data in a simple algorithm (that may easily be developed in conventional computer-program spreadsheets), appropriate for their use by medical personnel. Interpreting graphic results and quantitative outputs may aid in decision-making process in order to find the most effective treatment for every patient.
Acknowledgments This research has been supported by FONDECYT 1150899, the Programa Semilla, Universidad Aut onoma del Estado de Mexico-Universidad Austral de Chile, grant numbers 3338/2012 FS, and FS-2012-01, and the support of the Regional Government of Los Rıos, Chile, grant numbers FIC-R 12-117.
References 1. Bielefeld KA, Amini-Nik S, Alman BA: Cutaneous wound healing: recruiting developmental pathways for regeneration. Cell Mol Life Sci. 2013;70:2059–2081.
2. Callaghan T, Khain E, Sander L, Ziff R: A stochastic model for wound healing. J Stat Phys. 2006;122:909–924. 3. Cukjati D, Rebersek S, Miklavcic D: A reliable method of determining wound healing rate. Med Biol Eng Comput. 2001;39:263–271. 4. Gardner SE, Frantz RA, Hillis SL, Blodgett TJ, Femino LM, Lehman SM: Volume measures using a digital image analysis system is reliable in diabetic foot ulcers. Wounds. 2012;24:146–151. 5. Kantor J, Margolis DJ: A multicentre study of percentage change in venous leg ulcer area as a prognostic index of healing at 24 weeks. Br J Dermatol. 2000;142:960–964. 6. Ajmera I, Swat M, Laibe C, Novere NL, Chelliah V: The impact of mathematical modeling on the understanding of diabetes and related complications. CPT Pharmacometrics Syst Pharmacol. 2013;2:e54. 7. Arciero JC, Mi Q, Branca MF, Hackam DJ, Swigon D: Continuum model of collective cell migration in wound healing and colony expansion. Biophys J. 2011;100:535–543. 8. Gilman T: Wound outcomes: the utility of surface measures. Int J Low Extrem Wounds. 2004;3:125–132. 9. Robson MC, Hill DP, Woodske ME, Steed DL: Wound healing trajectories as predictors of effectiveness of therapeutic agents. Arch Surg. 2000;135:773–777. 10. Sherratt JA, Murray JD: Mathematical analysis of a basic model for epidermal wound healing. J Math Biol. 1991;29:389–404. 11. Sherratt JA, Murray JD: Epidermal wound healing: the clinical implications of a simple mathematical model. Cell Transplant. 1992;1:365–371. 12. Tranquillo RT, Murray JD: Mechanistic model of wound contraction. J Surg Res. 1993;55:233–247. 13. Wallenstein S, Brem H: Statistical analysis of wound-healing rates for pressure ulcers. Am J Surg. 2004;188(1A suppl l):73–78. 14. Gorin DR, Cordts PR, LaMorte WW, Menzoian JO: The influence of wound geometry on the measurement of wound healing rates in clinical trials. J Vasc Surg. 1996;23:524–528. 15. Hill DP, Poore S, Wilson J, Robson MC, Cherry GW: Initial healing rates of venous ulcers: are they useful as predictors of healing? Am J Surg. 2004;188(1A suppl l):22–25. 16. Margolis DJ, Gross EA, Wood CR, Lazarus GS: Planimetric rate of healing in venous ulcers of the leg treated with pressure bandage and hydrocolloid dressing. J Am Acad Dermatol. 1993;28:418–421. 17. Santamaria N, Ogce F, Gorelik A: Healing rate calculation in the diabetic foot ulcer: comparing different methods. Wound Repair Regen. 2012;20:786–789. 18. De Franciscis S, Fragomeni G, Caruso MV, et al: A new photographic computerized measurement system for chronic wound assessment. Acta Phlebol. 2014;15:13–18. 19. Barone S, Paoli A, Razionale AV: Assessment of chronic wounds by three-dimensional optical imaging based on integrating geometrical, chromatic, and thermal data. Proc Inst Mech Eng H. 2011;225:181–193. 20. Thawer HA, Houghton PE, Woodbury MG, Keast D, Campbell K: A comparison of computer-assisted and manual wound size measurement. Ostomy Wound Manage. 2002;48:46–53. 21. Little C, McDonald J, Jenkins MG, McCarron P: An overview of techniques used to measure wound area and volume. J Wound Care. 2009; 18:250–253. 22. Vidal A, Giacaman A, Oyarzun-Ampuero FA, et al: Therapeutic potential of a low-cost device for wound healing: a study of three cases of healing after lower-extremity amputation in patients with diabetes. Am J Ther. 2013;20:394–398. 23. Brandt MG, Moore CC, Micomonaco D, et al: A prospective randomized evaluation of scar assessment measures. Laryngoscope. 2009;119:841–845. 24. Jessup RL: What is the best method for assessing the rate of wound healing? A comparison of 3 mathematical formulas. Adv Skin Wound Care. 2006;19:138–147. 25. Vercelli S, Ferriero G, Sartorio F, Cisari C, Bravini E: Clinimetric properties and clinical utility in rehabilitation of postsurgical scar rating scales: a systematic review. Int J Rehabil Res. 2015;38:279–286. 26. Zimny S, Schatz H, Pfohl M: Determinants and estimation of healing times in diabetic foot ulcers. J Diabet Complications. 2002;16:327–332.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334
1 2 3 4 5 6 7 8 Q1 9 10 11 12 Q4 13 14 15 16 17 Q3 18 19 20 21 22 23 24 25 26 27 28 29 30 31 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 57 58 59 60 61
RESEARCH PAPER
A Simple Mathematical Model for Wound Closure Evaluation nb, Israel Giacamana, Alejandra Vidal Villaa, Hugo Mendieta Zero pezb, Laura E. Meza Trillod, Marıa del Socorro Camarillo Romeroc, Sandra Parra Lo David A. Perez P ereze, Miguel Conchaa, Cesar Torres-Gallegosf, Sandra L. Orellanaf, n-Ampuerog, Ignacio Moreno-Villosladag,* Felipe Oyarzu a
Instituto de Anatomıa, Histologıa y Patologıa, Facultad de Medicina, Universidad Austral de Chile, Valdivia, Chile Facultad de Medicina, Universidad Aut onoma del Estado de Mexico, Toluca, Estado de Mexico, Mexico c Centro de Investigaci on en Ciencias Medicas (CICMED), Universidad Autonoma del Estado de Mexico, Toluca, Estado de Mexico, Mexico d Facultad de Ciencias de la Salud, Universidad de Anahuac, Huixquilucan, Estado de Mexico, Mexico e Universidad del Valle de Toluca, Toluca, Estado de Mexico, Mexico f Instituto de Ciencias Quımicas, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile; and g Departamento de Ciencias & Tecnologıas Farmaceuticas, Universidad de Chile, Santiago, Chile b
KEYWORDS: Diabetic foot; Foot ulcers; Mathematical model; Wounds
Abstract The incidence of ulcers associated to type 2 diabetes mellitus (T2DM) increases every year. We introduce and explore a new mathematical algorithm to evaluate wound-healing in foot ulcers associated to T2DM. Fifteen patients (nine women and six men), mean age of 70 6 16 years were included. The evolution of their wounds followed-up for a period of 18–45 days. According to the Wagner grading system the ulcers were grade I (5 patients), grade II (9 patients), and grade III (1 patient). Clinically, the type of the ulcers was neuroischemic (12 patients) and neuropathic (3 patients). A new parameter is introduced, the ‘‘continuous linear healing rate’’ Dc that was more accurate with higher values and requires less quantifications than usual formulas to make a wound-healing projection. 2016 Elsevier Inc. All rights reserved.
Introduction Wound-healing is a complex biological process that involves the expression of growth factors that promote Conflict of interest/financial disclosure statement: Authors have nothing to declare. * Corresponding author. Instituto de Ciencias Quımicas, Facultad de Ciencias, Universidad Austral de Chile, Isla Teja, Casilla 567, Valdivia, Chile. Fax: 156 63 2293520. E-mail address: [email protected]
various cellular processes, production of new connective tissue matrix, and collagen deposition.1 Studies have shown that it is possible, under appropriate treatments, to achieve wound closure in about 45 days. Decision-making by nurses and physicians during wound-healing may be difficult due to lack of quantitative criteria which allow an accurate diagnosis of the quality of the wound progression. With the presence of a pool of protocols and products possessing the potential to improve wound progression to closure, the selection of one of these may become arbitrary. In order to perform quantitative
2213-5103/$ - see front matter 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jccw.2016.07.002
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
2 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195
Journal of the American College of Clinical Wound Specialists, Vol -, No -
analyses, data acquisition must be carried out, which adds another difficulty to the work of the clinical staff. Data acquisition should be simple and fast to be easily incorporated into healing protocols by nurses. In addition, data treatment and interpretation must not involve complicated protocols; either demand high qualification in disciplines other than medical, such as mathematics, physics, or computation. Stochastic models2 that consider delayed exponential behavior,3 or photogrammetric software,4 furnish reliable methods on the evaluation of diabetic foot ulcers, but the protocols, in terms of algorithms and the software required, may result sufficiently complicated to minimize their use by nurses in hospitals and family health centers, where treatments are normally performed. Evidence showed that monitoring healing progression for 2 or 4 weeks may serve to predict eventual wound closure under appropriate mathematical models.5 Mathematical modeling may also aid in identifying potential methods for disease control and management if wound closure results can be correlated with the wound and the patient’s condition.6 Few mathematical models in the literature deal with the complications associated with diabetes due to their complexity.6 The literature reveals great diversity in the procedures employed to compare wound-healing rates, including variables such as periods of study, type, size and shape of the analyzed lesions, closing rate definition, calculation methods, mechanisms, etc.3,7–13 However, controversy exists in the selection of the adequate indicative parameter of the wound closure rate.8,9,14,15 Assessment of wound status in terms of time is often based on measurements of wound area and, to a lesser extent, of the wound perimeter8,14–17 and the linear healing rate (D) has been often utilized. The D parameter is obtained through a method of calculation based on differences; thus, accurate calculation requires a large number of data to achieve statistical significance. However, the concept of linear healing involves a constant, perpendicular advance of the wound margin toward the wound’s open area, and this feature may be used to propose a mathematical development based on differentials. The aim of this study is to introduce and test a mathematical algorithm for calculation of wound-healing rate from few clinical data obtained by monitoring wound closure in patients with Type 2 Diabetes Mellitus (T2DM).
Materials and Methods The research study protocol was approved by the Ethics Committee of the Medical Sciences Research Center (CICMED), Autonomous University of the State of Mexico (UAEMex). The investigation conforms to the principles in the Declaration of Helsinki, and the patients were fully informed regarding the aim of the study and signed written informed consent.
This was a bi-center, open-label, phase I, descriptive study. A total of 15 consecutive patients with T2DM and foot ulcer up to grade III on the Wagner scale were enrolled in this study. Patients were recruited at the Healing Ulcer Service, Clınica 251, Instituto Mexicano del Seguro Social (IMSS), Metepec, Estado de Mexico (3 patients), and at the Clınica de Consulta Externa ‘‘Alfredo del Mazo Velez,’’ ISSEMyM, Toluca, Estado de Mexico (12 patients) between September 2012 and August 2013. Other subjects who had clinical signs of severe infection and those unwilling to participate in the study were excluded.
Clinical Follow-up Age, gender, duration of diabetes, time of wound evolution, Wagner ulcer grade, and vascular status were recorded for all patients. The ulcer was delineated for each patient on a sterile acetate sheet, and a scale was added. Images were digitized and analyzed with AutoCAD 2009 software to determine their area and perimeter. Five or more measurements were carried out at the beginning of the study and during follow-up for 18–45 days. Patients were advised to have their dressing changed daily, either at the outpatient clinic of our hospital, at a nearby local hospital, or with the aid of the community nurse who provided home care for patients with ambulatory problems.
Mathematical Model According to Gilman,8 we can define the linear healing rate (D) as follows: D52
DA Pavg Dt
ð1Þ
where DA is the change in superficial wound area between 2 consecutive clinic visits, Dt is the time between the two visits, and Pavg is the average wound perimeter measured at the two clinic visits. This equation is evaluated in discrete times during the wound-healing process, normally once a week, and the final value is averaged among the discrete values obtained at each discrete time, obtaining the average D value (Davg). The drawback of this method is that in order to obtain statistical significance, a large number of visits and data acquisitions are needed, and data frequently exhibit great dispersion. However, the concept of linear healing involves a regular advance of the wound contour toward the wound open area and perpendicular to the wound margin; thus, data corresponding to this magnitude are theoretically linked with each other by a linear trend over time. This concept may be mathematically formulated by considering the effective advance of the wound margin during an infinitely small period of time (Fig. 1). Thus, the continuous linear healing rate (Dc) can be defined as:
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262
Villa et al 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
---
Dc 5 lim D5 lim 2 Dt/0
Dt/0
3 DA dA dr 52 52 Pavg Dt Pdt dt
ð2Þ
where A: area, d: delta, lim: limit, P: perimeter, Pavg: average wound perimeter measured at two clinic visits, r: length perpendicular to the wound contour, DA: change in superficial wound area between 2 consecutive clinic visits, Dt: time between two visits. dA 5 lim DA when DA approaches 0. Considering that according to Fig. 1: ð3Þ
dA5Pdr
where A is superficial wound area, r is the length perpendicular to the wound contour, and P is the wound perimeter. Then; it is clear that dA52PDc dt
ð4Þ
Integration of equation (4) leads to A 52Dc t1q P
P2 A
tclosure 5
q Dc
ð7Þ
If we consider linear healing as a physiological fact, deviation from the linear trend following equation (5) should be caused by changes in the shape of the wound and other physiological conditions that may influence the value of Dc during wound-healing.
Statistical Analysis ð5Þ
where q is the intercept, denoting the extrapolated A/P initial values. Plotting A/P vs. time, the values of Dc and q can be obtained from the slope and the ordinate at the origin, respectively. Basically, q contains information on the aspect of the wound, such as those comprising extrapolated A/P initial values. Because P and A have dimensions of length and square length, respectively, we can define the Shape coefficient (Sc) as: Sc 5
This coefficient does not depend on the size of the analyzed figure, but on its shape, and may be an appropriate parameter for evaluating changes in the shape of a particular wound or differences among the shapes of different wounds. This assumes a value of 12.56 for circular wounds and 16 for square wounds, as depicted in Fig. 1. In addition, equation (5) can be used to obtain projected time of wound closure (tclosure), a condition under which A 5 0, so that:
ð6Þ
STATISTICA version 8.0 software was utilized to analyze statistical correlations between variables; statistical differences by grouping patients by Wagner’s grade of wound type were searched using the Mann–Whitney U test.
Results Fifteen patients (nine females and six males) with a mean age of 70 6 16 years were included in the study. According to the Wagner grading system, ulcers were grade I (5 patients), grade II (9 patients), and grade III (1 patient).
Figure 1 Wound evolution scheme for different shapes. A: area, Dc: continuous linear healing rate, lim: limit, P: perimeter, Pavg: average wound perimeter measured at two clinic visits, r: length perpendicular to the wound contour, Sc: shape coefficient, DA: change in superficial wound area between 2 consecutive clinic visits, Dt: time between two visits.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
Q2
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396
Journal of the American College of Clinical Wound Specialists, Vol -, No -
4 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
Table 1
Anthropometric and clinical characteristics of the patients.
Case
Gender
Age
Duration of diabetes (years)
Time with the ulcer (months)
Wagner stage
Wound type
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
M F F F M M F M F F M F M F F
56 62 79 91 85 71 40 74 97 45 75 62 67 66 81
25 4 5 26 35 6 13 14 18 1 5 1.6 5 2 8
12 2 2 3 2 6 63 4 5 6 8 2 3 1 5
II II II II II III II I I II II I I I II
Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuroischemic Neuropathic Neuroischemic Neuroischemic Neuroischemic Neuropathic Neuroischemic Neuropathic Neuroischemic
F: female, M: male, P: patient.
Clinically, the ulcer type was neuroischemic (12 patients) and neuropathic (3 patients). Duration of ulcer evolution prior to treatment ranged between 1 and 63 months (Table 1).
Ulcer Healing Data obtained from planimetric measurements have been treated introducing A, P, and t in equations (1), (5) and (6). D values obtained from equation (1) for the healing process of the 15 patients are shown in Fig. 2, as they are usually presented in the literature.15 The corresponding Davg values are listed in Table 2, together with the corresponding values of the absolute and relative Standard Deviations (SD and RSD, respectively). D values of the same order of magnitude as those
Figure 2
usually reported were found.17 However, in the majority of cases, corresponding RSD were .100% (Table 2), with an outstanding value close to 800% for patient (P) 14. RSD was used to evaluate the quality of the data for the Davg calculation, and the high values found indicate low quality. In addition, there was no correlation between the Davg value and RSD values (r 5 0.4834; p 5 0.68); therefore, the data dispersion for each patient was assumed to be independent from the healing rate. High dispersion of D values for each single patient makes it difficult to establish comparisons between patients. On the other hand, utilizing the same raw data, Dc values have been obtained by means of equation (5) and are illustrated in Table 3. In order to obtain trends deriving from the same origin, A/P data and the corresponding linear adjustments have been plotted in Fig. 3 for all patients after
D (?) for each patient. D: linear healing rate.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530
Villa et al 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597
Table 2
---
5
Davg and corresponding SD and RSD.
Patient
Davg 6 SD (cm/day)
RSD (%)
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
0.00828 6 0.01424 20.00701 6 0.03318 0.00939 6 0.01858 20.00868 6 0.02981 0.00916 6 0.00871 0.01094 6 0.00959 0.00263 6 0.00453 0.00740 6 0.01024 0.00580 6 0.01060 0.00795 6 0.01657 0.00244 6 0.00656 0.01012 6 0.00737 0.00484 6 0.00623 0.00153 6 0.01222 0.00401 6 0.00535
172 473 198 344 95 88 172 138 183 208 269 73 129 796 133
Davg: average linear healing value, RSD: relative standard deviation, SD: standard deviation.
subtracting q. The resulting straight lines were distributed in a cone. Expressions of linear adjustments of A/P vs. t and the corresponding R2 values are also presented in Table 3. R2 values were also used to evaluate data quality for calculation of Dc, based on the assumption of a linear behavior. R2 values found presented values between 0 and 1. There was a positive correlation between Dc and R2 (r 5 0.6384; p 5 0.010). Because this correlation was positive and statistically significant, we can conclude that higher Dc values resulted more accurate than lower values. One patient (P2) exhibited negative values of Dc, indicating that the wound was enlarging, and a very low value of R2
Table 3
was found (0.03). Six patients showed positive Dc values of ,0.002 cm/day (P11, P9, P15, P14, P7, and P13), and the corresponding R2 ranged between 0.22 and 0.60. Six patients demonstrated values of around 0.004 cm/day (P8, P12, P4, P1, P3, and P6), and the corresponding R2 ranged between 0.24 and 0.96. Two patients showed Dc values of .0.006 cm/day (P10 and P5), and the corresponding R2 were 0.62 and 0.95. Total wound closure should occur at tclosure (extracted from equation (7)) whose values are shown in Table 3 for patients showing Dc values. The values obtained for tclosure ranged between 52 and 805 days. The mean tclosure value obtained for the patients was 260 days, and the data showed an SD of 212 (RSD, 81%). In order to compare the results obtained from two different methods, Dc and Davg values were plotted in Fig. 4a, where patient entries were ordered according to Dc values. In that Dc and Davg were calculated from the same data, a positive correlation was expected between both magnitudes, which was statistically confirmed (r 5 0.5490; p 5 0.034). However, outstanding differences were found in P2 and P4. For P4, an inversion on the sign of the corresponding value was observed, because Dc was positive and Davg was negative. In general, positive values of both magnitudes were found, with Dc lower than Davg. In addition, there was no correlation between the corresponding R2 (corresponding to the calculation of Dc by linear adjustment) and RSD corresponding to the calculation of Davg (r 5 0.2267; p 5 0.417) (Tables 2 and 3). Notably, for patients exhibiting high values of R2 when calculating Dc (.0.90), such as P1, P5, and P12, the RSD values found were still high and close to 100%. Assays for observing any correlation between Dc or Davg with wound shape and size have been conducted. In this
Linear adjustments of A/P data versus t, Dc, Sc, and corresponding SD and RSD.
Patient
Linear adjustment (equation (6))
R2
Dc (cm/day)
Sc 6 SD
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15
y y y y y y y y y y y y y y y
0.96 0.03 0.52 0.24 0.95 0.83 0.60 0.85 0.33 0.62 0.22 0.94 0.82 0.53 0.48
0.00396 20.00080 0.00438 0.00380 0.00686 0.00496 0.00152 0.00351 0.00069 0.00660 0.00034 0.00365 0.00189 0.00010 0.00095
13.74 14.86 13.40 18.54 18.65 14.54 20.23 14.04 15.43 15.64 17.01 15.90 15.10 14.35 14.60
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
20.00396x 1 1.25068 0.00080x 1 0.38320 20.00438x 1 0.65355 20.0038x 1 0.40481 20.00686x 1 0.35369 20.00496x 1 0.68489 20.00152x 1 0.24996 20.00351x 1 0.44695 20.00069x 1 0.22981 20.00660x 1 1.12439 20.00034x 1 0.21049 20.00365x 1 0.53096 20.00189x 1 0.36564 20.00010x 1 0.32471 20.00095x 1 0.76603
6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
0.34 0.89 0.34 0.77 5.35 0.22 2.48 0.31 0.94 0.84 0.48 0.73 0.52 0.22 0.40
RSD (%)
tclosure (days)
2.47 5.96 2.53 4.17 28.7 1.49 12.3 2.21 6.08 5.38 2.80 4.61 3.45 1.52 2.71
316 – 149 106 52 138 165 127 331 170 622 145 193 326 805
Dc: linear healing value, RSD: relative standard deviation, Sc: shape coefficient, SD: standard deviation, tclosure: time of wound closure, y 5 A/P (cm), x 5 t (days), R2: linear regression factor.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664
665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731
Journal of the American College of Clinical Wound Specialists, Vol -, No -
print & web 4C=FPO
6
Figure 3 (A/P) – q as a function of time. Data points and linear arrangements for the 15 patients. A: area, P: perimeter, q: intercept of the formula A/P 5 2Dct 1 q, where Dc: continuous linear healing rate and t: time (days).
regard, four magnitudes related with initial wound shapes and sizes have been evaluated as follows: initial Area (Ainit); initial Perimeter (Pinit ); initial Shape coefficient (Scinit) init init defined as Sc 5 (P )2/Ainit, and q. The respective values were plotted in Fig. 4b. Ainit ranged between 0.78 and 21.1 cm2, and no correlation with either Dc (r 5 0.5015; p 5 0.057) or Davg (r 5 0.2311; p 5 0.407) was found. Pinit ranged from 3.4 to 17 cm, and respective values correlated with Dc (r 5 r 20.5376; p 5 0.039), but did not correlate with Davg (5 0.1687; p 5 0.548). A negative correlation was found between Scinit and Davg (r 5 20.6101; p 5 0.016), but not between Scinit and Dc (r 5 20.3361; p 5 0.221). q values ranged between 0.21 and 1.25 cm, demonstrating no correlation with Dc (r 5 20.353; p 5 0.197) or Davg (r 5 20.272; p 5 0.326). Thus, taking these results as a whole, we can conclude that there is a weak correlation between the healing rates found by both mathematical methods and the initial wound shape and size. Wound shape evolution during this study has been documented by the Sc calculation through equation (6). As previously mentioned, Sc is a parameter that does not depend on the size of the analyzed figure, but on its shape. Likewise, a perfect circle has a Sc of 4p, the lowest possible value to be found (Fig. 1). Sc mean values, together with the corresponding SD and RSD, are plotted in Fig. 5a. In this study, Sc values ranged between 13.1 and 18.2, indicating that wounds were not elongated toward any specific direction. There was no correlation (r 5 0.227; p 5 0.417) between R2 values of the linear adjustments (Table 3) from which Dc was calculated and RSD obtained for each patient during wound evolution, indicating that a change in wound shape may not be a cause of loss of linearity. Similarly, no correlation was found between the RSD values from which
Davg was calculated and RSD (r 5 20.231; p 5 0.408), indicating that data dispersion for any patient was not necessarily caused by a change in wound shape during the healing process. In fact, small changes in wound shape have been registered, a witnessed by low values of RSD, .6%, with the exception of P5 and P7, for which the corresponding RSD yielded values of 28.7 and 12.3%. Evolution in time of this parameter for these two patients can be observed in Fig. 5b. The wound shape for P5 clearly changed after the first 15 days, adopting a more elongated form during the last 2 weeks (Fig. 5c). Evolution of the wound of P7 appears more erratic, demonstrating Sc values trending up and down, as can be noted in Fig. 5b and c. However, evolution of the wound of P5 exhibited excellent linear behavior in the wound-healing process, with an R2 value of 0.95 for linear adjustment of data, whereas an R2 of only 0.60 was obtained for P7. In addition, data for P5 are less dispersed than those for P7 when calculating Davg, demonstrating RSD of 95 and 172%, respectively (Table 1). We also noted that the patient’s health condition reported in Table 1 could not be correlated with Dc or Davg. Furthermore, no statistical differences were found considering Wagner grade (Table 1) and Dc (U 5 17; p 5 0.463) or Davg (U 5 21; p 5 0.841), employing the Mann–Whitney U test. For this test, the sample corresponding to P6, the only Wagner III patient, has not been considered. Similarly, no differences were found regarding wound type (Table 1) and Dc (U 5 16; p 5 0.7728) or Davg (U 5 16; p 5 0.7728), utilizing the same statistical test. In fact, only two of the three neuropathic patients showed Dc values . 0.004 cm/day (P8 and P12). The only Wagner III patient (P6) presented a Dc value of 0.00496 cm/day. Wagner I ulcers corresponded to the three neuropathic patients (P8, P12, and P14) plus P9 and P13, and all patients presented Dc values , 0.004 cm/day. The two younger patients (P7 and P10, aged 40 and 45 years, respectively), showed neuroischemic ulcers, with similar Dc values of around 0.006, despite the duration of both of their wounds and that the pathologies differed from each other.
Discussion The integrated of data acquisition technology for the measurement of wound extension, such as photographic systems with adequate software to record and analyze images, and systems that employ stereophotometry, may furnish additional advantages, such as low interobserver variability. In addition, sophisticated systems are able to distinguish by color analysis areas with granulation and necrotic tissue.18,19 After image treatment, parameters such as A and P may be obtained at different time values. In classical, simple analyses, affordable by nurses and physicians at hospitals and family health centers, healing rates are obtained by calculation of absolute and relative surface
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865
---
7
print & web 4C=FPO
Villa et al
Figure 4 a) Dc (-) and Davg ( ) for each patient; b) Ainit (-), Scinit (,), q values (-), and Pinit ( ) for each patient. Ainit: initial area, Davg: average linear healing value, Dc: continuous linear healing rate, Pinit: initial perimeter, q: intercept of the formula A/P 5 2Dct 1 q, where Dc: continuous linear healing rate and t: time (days), Scinit: initial shape coefficient.
changes in time and calculation of parameter D.17 However, these analyses may fail due to large variations in D values obtained between two consecutive clinic visits. Thus, as what has been shown in this study, the high data dispersion achieved when calculating Davg would not allow for extracting useful information for decision-making with respect to changing treatment, for example, or for applying any specific product. In our approach, considering the ease of data acquisition, planimetry of the wounds on acetate sheets represents a simple and short time-saving strategy (Fig. 6).20,21 The alternative data treatment developed here permits the extraction of hidden information, based on the assumption implicit in the definition of linear healing, i.e., a regular advance of the wound contour toward the wound open area and perpendicular to the wound margin, so that data corresponding to this magnitude are theoretically linked with each other by a linear trend over time. Thus,
the slope of the linear trend, which we denominate Dc, arises from the introduced mathematical model to characterize wound-healing. In addition, the Dc parameter possesses an advantage over Davg that allows for the use of a lower number of values for the analyses, in that the data are linked by a mathematical formula. Also, it provides better visual information because the parameter is represented by the slopes of straight lines in a graph where the wound evolution for different patients can be compared. In order to better understand these terms, the data obtained for P5 are presented in Fig. 7. In treating data according to the classical method, therefore calculating Davg by means of equation (1), the differences in A and P between two consecutive visits, independent from differences between two other consecutive visits, must be considered. Moreover, individual D values are classically calculated only by treating the data, but
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932
Journal of the American College of Clinical Wound Specialists, Vol -, No -
8
Figure 5 a) Mean Sc ( ) (with SD) and respective RSD (-) for each patient; b) Sc values for P5 (?) and P7 (-); c) Wound margins for P5 and P7 (out of scale). P (number): number of the patient, RSD: relative standard deviation, Sc: shape coefficient, SD: standard deviation.
not graphically. We represent the A/P data vs. t in Fig. 7 (top left and right). The individual D values obtained between each pair consecutive visits are represented by the
slopes of the small segments (Fig. 7, top left). The lines do not completely adjust to the segments linking two consecutive data, in that equation (1) employs Pavg instead
print & web 4C=FPO
933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999
Figure 6
A) Photograph and B) its scanned image of one case of foot ulcer.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066
Villa et al 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133
---
9
Figure 7 Top left: Data points obtained for P5 (?), and its treatment according to equation (1); individual D values are graphically shown as the slope of the lines. Bottom left: individual D values are plotted (C), together with the corresponding Davg (-) and its RSD. Top right: Data points obtained for P5 (?), and their linear adjustment according to equation (5). Bottom right: the Dc value and the corresponding R2 obtained are shown. A: area, D: linear healing rate, Davg: average linear healing value, P: perimeter, R2: linear regression factor, RSD: relative standard deviation.
of P. Then, a large dispersion in D values is found, since the trending up and down of the data produces large differences in the slope of the corresponding segments; thus, RSD values are also found to be large (Fig. 7, top and bottom left). Expression (5) has the advantage over expression (1) in that fewer data points are needed to trace the healing profile, because of the theoretical linear trend; therefore, linkage between data, is considered. Thus, from the mathematical point of view, five visits taken once a week where data are acquired should be sufficient to find the linear trend of the wound-healing by fitting five experimental points. The innovative method presented reflects that the data are not independent one from the other, but are linked by a linear trend. Thus, Dc can be easily calculated, with few data, by obtaining the slope of this linear trend after linear adjustment, with R2 as the parameter for evaluating the strength of the linear association among the variables. Note that, for P5, the data followed a linear trend with an excellent R2 of 0.95, which confirmed the good quality of the Dc estimation (Fig. 7, right), whereas an RSD of 95% was obtained for the Davg calculation.
Among information obtained by introducing data on the mathematical algorithm, is the tclosure, a parameter that could be easily taken into consideration to make a clinical decision. For example, it has been reported, for diabetic foot ulcers that, linearity of the healing process may be lost for ulcers healing beyond 28 days.17,22 In this regard, tclosure should be interpreted as an extrapolation of linear healing during the first 28–35 days of healing, and used for a decision to change treatment rather than as a predictive value. In addition, the correlation found between the values of Dc and R2 may supply another criterion for decision-making. Low Dc values with low R2 may indicate poor quality of wound-healing progression. On the contrary, high Dc values with high R2 may indicate that the treatment is effective. For example, if Dc values . 0.003 cm/day and R2 .0.8 are imposed as quality criteria (to this point, arbitrarily), 5 patients (P1, P5, P6, P8, and P12) comply with this requisite. Moreover, these parameters offer means of comparison of the wound evolution of a specific patient before and after changing the treatment, once the latter is decided upon by the medical personnel.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
10 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267
Journal of the American College of Clinical Wound Specialists, Vol -, No -
In clinical setting, mathematical formulas based on the argument that the dimension under assessment conforms to a linear phenomenon23 have been used not only to evaluate wound-healing24 but also scar evolution.25 The data obtained from this small sample of patients suggest that we can predict the type of response to the healing strategy after 4–6 weeks of follow-up. If this is corroborated with a larger study, the possibilities to replicate the prediction of this new formula in other diseases are high, more if we add the use of complex image studies like axial computed tomography (ACT) or Magnetic Resonance Image (MRI). Data accumulation will aid in establishing quantitative criteria for making a diagnosis of the quality of wound progression and for obtaining more accurate parameters and criteria for comparison of the wound-healing evolution for different patients. The incorporation in the future of more technological tools for image capture, processing, and analysis in clinical practice may be slow, but probably unavoidable. It has been reported in the literature that, providing standard care, the time course of wound-healing in diabetic foot ulcers is predominantly determined by etiologic factors, and to a lesser degree by wound size.26 However, both the patient’s health condition and the wound shape have been invoked to explain differences in healing rates.14,15,17 In this study, null or weak correlations have been found between Dc or Davg in terms of wound size and shape, parameterized by means of A, P, and Sc, or with the patient’s condition. A clear limitation of the study is the low number of patients. In spite of this drawback, this was a pilot study and the advantages provided by the calculation of Dc and related parameters may represent an important feature for the medical management of several wound types, because accurate results may be easily obtained solely by introducing a few data in a simple algorithm (that may easily be developed in conventional computer-program spreadsheets), appropriate for their use by medical personnel. Interpreting graphic results and quantitative outputs may aid in decision-making process in order to find the most effective treatment for every patient.
Acknowledgments This research has been supported by FONDECYT 1150899, the Programa Semilla, Universidad Aut onoma del Estado de Mexico-Universidad Austral de Chile, grant numbers 3338/2012 FS, and FS-2012-01, and the support of the Regional Government of Los Rıos, Chile, grant numbers FIC-R 12-117.
References 1. Bielefeld KA, Amini-Nik S, Alman BA: Cutaneous wound healing: recruiting developmental pathways for regeneration. Cell Mol Life Sci. 2013;70:2059–2081.
2. Callaghan T, Khain E, Sander L, Ziff R: A stochastic model for wound healing. J Stat Phys. 2006;122:909–924. 3. Cukjati D, Rebersek S, Miklavcic D: A reliable method of determining wound healing rate. Med Biol Eng Comput. 2001;39:263–271. 4. Gardner SE, Frantz RA, Hillis SL, Blodgett TJ, Femino LM, Lehman SM: Volume measures using a digital image analysis system is reliable in diabetic foot ulcers. Wounds. 2012;24:146–151. 5. Kantor J, Margolis DJ: A multicentre study of percentage change in venous leg ulcer area as a prognostic index of healing at 24 weeks. Br J Dermatol. 2000;142:960–964. 6. Ajmera I, Swat M, Laibe C, Novere NL, Chelliah V: The impact of mathematical modeling on the understanding of diabetes and related complications. CPT Pharmacometrics Syst Pharmacol. 2013;2:e54. 7. Arciero JC, Mi Q, Branca MF, Hackam DJ, Swigon D: Continuum model of collective cell migration in wound healing and colony expansion. Biophys J. 2011;100:535–543. 8. Gilman T: Wound outcomes: the utility of surface measures. Int J Low Extrem Wounds. 2004;3:125–132. 9. Robson MC, Hill DP, Woodske ME, Steed DL: Wound healing trajectories as predictors of effectiveness of therapeutic agents. Arch Surg. 2000;135:773–777. 10. Sherratt JA, Murray JD: Mathematical analysis of a basic model for epidermal wound healing. J Math Biol. 1991;29:389–404. 11. Sherratt JA, Murray JD: Epidermal wound healing: the clinical implications of a simple mathematical model. Cell Transplant. 1992;1:365–371. 12. Tranquillo RT, Murray JD: Mechanistic model of wound contraction. J Surg Res. 1993;55:233–247. 13. Wallenstein S, Brem H: Statistical analysis of wound-healing rates for pressure ulcers. Am J Surg. 2004;188(1A suppl l):73–78. 14. Gorin DR, Cordts PR, LaMorte WW, Menzoian JO: The influence of wound geometry on the measurement of wound healing rates in clinical trials. J Vasc Surg. 1996;23:524–528. 15. Hill DP, Poore S, Wilson J, Robson MC, Cherry GW: Initial healing rates of venous ulcers: are they useful as predictors of healing? Am J Surg. 2004;188(1A suppl l):22–25. 16. Margolis DJ, Gross EA, Wood CR, Lazarus GS: Planimetric rate of healing in venous ulcers of the leg treated with pressure bandage and hydrocolloid dressing. J Am Acad Dermatol. 1993;28:418–421. 17. Santamaria N, Ogce F, Gorelik A: Healing rate calculation in the diabetic foot ulcer: comparing different methods. Wound Repair Regen. 2012;20:786–789. 18. De Franciscis S, Fragomeni G, Caruso MV, et al: A new photographic computerized measurement system for chronic wound assessment. Acta Phlebol. 2014;15:13–18. 19. Barone S, Paoli A, Razionale AV: Assessment of chronic wounds by three-dimensional optical imaging based on integrating geometrical, chromatic, and thermal data. Proc Inst Mech Eng H. 2011;225:181–193. 20. Thawer HA, Houghton PE, Woodbury MG, Keast D, Campbell K: A comparison of computer-assisted and manual wound size measurement. Ostomy Wound Manage. 2002;48:46–53. 21. Little C, McDonald J, Jenkins MG, McCarron P: An overview of techniques used to measure wound area and volume. J Wound Care. 2009; 18:250–253. 22. Vidal A, Giacaman A, Oyarzun-Ampuero FA, et al: Therapeutic potential of a low-cost device for wound healing: a study of three cases of healing after lower-extremity amputation in patients with diabetes. Am J Ther. 2013;20:394–398. 23. Brandt MG, Moore CC, Micomonaco D, et al: A prospective randomized evaluation of scar assessment measures. Laryngoscope. 2009;119:841–845. 24. Jessup RL: What is the best method for assessing the rate of wound healing? A comparison of 3 mathematical formulas. Adv Skin Wound Care. 2006;19:138–147. 25. Vercelli S, Ferriero G, Sartorio F, Cisari C, Bravini E: Clinimetric properties and clinical utility in rehabilitation of postsurgical scar rating scales: a systematic review. Int J Rehabil Res. 2015;38:279–286. 26. Zimny S, Schatz H, Pfohl M: Determinants and estimation of healing times in diabetic foot ulcers. J Diabet Complications. 2002;16:327–332.
FLA 5.4.0 DTD - JCCW59_proof - 12 August 2016 - 7:24 pm - ce
1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334