<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>The time factor as an associative concept relation in modelling post-liver transplant management complications</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Paul Sambre</string-name>
          <email>paul.sambre@ku-</email>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Cornelia Wermuth</string-name>
          <email>cornelia.wermuth@ku-</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Hendrik J. Kockaert</string-name>
          <email>hendrik.kockaert@ku-</email>
          <xref ref-type="aff" rid="aff2">2</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>MIDI</institution>
          ,
          <addr-line>Multimodality, interaction and, discourse</addr-line>
          ,
          <institution>University of Leuven</institution>
          ,
          <country country="BE">Belgium</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>MIDI</institution>
          ,
          <addr-line>Multimodality, interaction, and discourse</addr-line>
          ,
          <institution>University of Leuven</institution>
          ,
          <country country="BE">Belgium</country>
        </aff>
        <aff id="aff2">
          <label>2</label>
          <institution>QLVL</institution>
          ,
          <addr-line>Quantitative lexicology and, variational linguistics</addr-line>
          ,
          <institution>University of Leuven, Belgium &amp; University of The, Free State</institution>
          ,
          <addr-line>South-Africa</addr-line>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2015</year>
      </pub-date>
      <fpage>107</fpage>
      <lpage>114</lpage>
      <abstract>
        <p />
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>We propose a first termontological analysis
of temporal parameters and relations applied
to the case of medical complications in
postliver transplant management (PLTM).
Medical complications contribute to different
degrees of morbidity and mortality in the
process of medical follow-up after transplant
surgery. Understanding the full ontological
and conceptual complexity of such complex
spanning (SPAN) time events time is a
central issue in drawing an implementable
semantic map of the potential causes of early
and long term complications, their diagnosis
and the potential effects due to medical
treatment. The analysis is usage-based and relies
on linguistic utterances for complications in
a concise medical review article.</p>
    </sec>
    <sec id="sec-2">
      <title>Intro: medical complications and associative relations in termontology</title>
      <p>
        This paper deals with medical complications and
their termontological description. Termontology
        <xref ref-type="bibr" rid="ref11">(Roche, 2007)</xref>
        combines insights from
terminology and applied ontology, combining them with a
linguistic dimension: whereas terminology is
basically interested in complex tree representations
between terms and their normalized definitions,
applied ontology models (visual) representations
of complex concept systems, including semantic
relations that dynamically connect static entities
in a terminological database. This paper combines
both preoccupations, starting from descriptions of
medical complications, more specifically those
occurring in post-liver-transplant management
(PLTM). In the area of liver transplantation,
termontology improves understanding of conceptual
structures based on the lexico-grammatical
structures retrieved from scientific literature about
complications following transplantation. So far,
medical complications have not been the subject
of much analysis in terminological or applied
ontology. Yet, they represent a rich resource for
investigating cause-effect relations that constitute a
specific subcase of causal events. In a broader
sense, a complication is any adverse, undesired
and unintentional result of disease management.
More technically, medical complications are
iatrogenic (i.e. disease-related) subsets of
cause-effect relations and refer to the (negative, adverse)
morbid consequence of a disease or disorder
resulting from unsuccessful disease management.
Complications, occasionally known as
suboptimal (negative) outcome, are not to be confused
with sequelae of previous acute medical
conditions and therapies. Negative outcomes can be
subdivided into failure to cure (pre-existing
conditions that remain unchanged after the
transplantation procedure), negative sequel and
complication
        <xref ref-type="bibr" rid="ref2">(Clavien et al., 2006)</xref>
        .
Complications differ from the commonly expected sequelae
resulting from an anterior acute medical condition
and therapy. Understanding and dealing
efficiently with postsurgical medical complications is
a crucial issue for the improvement of healthcare
quality, since complications require longer and
more expensive treatment, and, more importantly,
will negatively affect patient prognosis. In
extreme cases, they may lead to severe co-morbidity
and even death.
      </p>
      <p>
        Concept relations are a key concept in knowledge
representation, because they interconnect the
different concepts or entities in a given knowledge
domain. Despite extensive work on concept
relations
        <xref ref-type="bibr" rid="ref7">(e.g. Khoo and Jin-Cheon, 2006)</xref>
        , a
descriptive, usage-based account of associative relations
is lacking to date
        <xref ref-type="bibr" rid="ref12">(Sambre and Wermuth, 2010)</xref>
        .
In the termontological approach, the focus has
been on (static) hierarchical concept relations
such as type-token and/or meronymic relations.
      </p>
      <p>
        Biomedical terminology uses these vertical
relations for unique designations of medical
concepts and their terminological variants that result
in a compendium of several controlled
vocabularies such as the Unified Medical Language System
(UMLS) and the controlled thesaurus of Medical
Subject Headings (MeSH)
        <xref ref-type="bibr" rid="ref4">(Grabar et al., 2012)</xref>
        .
These classifications display conceptual and
socalled static snap shots of medical events.
Biomedical ontologies, conversely, aim to explore
snap shot relations against a dynamic background
of temporal unfolding, which results in so-called
SPAN (or spanning time) relations. This approach
reflects the true conditions as medical concepts
simultaneously express both hierarchical and
associative (i.e. time-based) relations. The
dynamicity of medical concepts is a phenomenon
that deserves further study. In this paper, we
therefore investigate in greater detail dynamic
associative relations, focusing on the temporal
underpinnings of causality in medical complications
in post-liver transplant management (PLTM). The
rationale is that causes, by definition, precede
effects both from a logical and experiential
perspective. Thus, the medical complications under
investigation can be assumed to entail both causal
and time-related relations. Our primary objective
is to set out temporal parameters to be used in a
conceptual model and to inventory specific
associative time elements inherent to the concept of
medical complication in PLTM.
2
      </p>
    </sec>
    <sec id="sec-3">
      <title>Time in the conceptual map of medical complications</title>
      <p>
        Time, in our view, is the underlying conceptual
basis or background against which causality of
(un)intentional medical events occur, as these
events are or are not triggered by instrumental
actions performed by medical teams. The following
example taken from a specialized journal on
transplantation in medicine should help to illustrate our
objective. The example discusses a complication
in PLTM, i.e. hepatic artery thrombosis (HAT):
“[…] hepatic artery thrombosis (HAT) is the
second main cause of liver graft failure. Moreover,
HAT is the most common vascular complication
in orthotopic liver transplantation (OLT). It is
associated with a marked increase in morbidity,
being the leading cause of graft loss (53%) and
mortality (33%) during the immediate
postoperative period. […] the consensus definition for early
HAT was an arterial thrombosis detected during
the first month after OLT. Late HAT was also
defined as the event detected ≥ 1 month after OLT.
[…] The true incidence of early HAT is unknown,
but it varies between 0% and 42%. […]
improvements in postoperative care have led to a marked
reduction in its incidence.”
        <xref ref-type="bibr" rid="ref10">(Pareja et al., 2010)</xref>
        As can be seen from the example, there is clear
linguistic evidence of both hierarchical and
associative relations. The following observations can
be made:
1. The linguistic expression (italics) designating
the most common vascular complication, the
hepatic arterial thrombosis, refers to a
hierarchical type of thrombosis (kind_of relation).
2. Causal information is provided by the
linguistic expressions cause of liver graft failure,
cause of graft loss and mortality.
3. Implicit instrumental reference is made to the
medical treatment in (complex) nouns such as
liver transplantation, graft failure and
postoperative care.
4. Temporal relations are set up between the
medical treatment and the post-surgery
adverse effect by means of prepositions
expressing a relation of time such as during the
immediate postoperative period, during the
first month after OLT, or ≥ 1 month after OLT.
From the above we can conclude that in medical
discourse recurrent, grammatically complex
linguistic patterns are used in order to connect (1) the
entities and types, (2) the causes and (unintended)
effects of health care, as well as (3) the
instrumental treatment administered by medical doctors
against a sequential background of (4) time. The
interplay between these four kinds of relations
may serve as a lexico-grammatical starting point
for drawing up a conceptual map of the medical
subdomain under investigation.
      </p>
      <p>Such conceptual maps are fundamental for
improving general procedures (in English) regarding
complications for medical care in
multidisciplinary medical treatment. PLTM occurs in hospital
teams, where physicians and nurses specialized in
complementary fields of haematology, radiology,
internal medicine, surgery and intensive care etc.,
collaborate in order to improve survival rates and
reduce the impact of complications.
Patient-centered healthcare implies collaborative settings that
call for efficient IT support systems to monitor
patient status and share information on the needs of
patients. These maps are to be shared between the
healthcare team’s actors, providing the scenarios
they share and the input of each team section or
member based on their individual knowledge
levels, their educational training, and the different
services and platforms these persons work in.
3</p>
    </sec>
    <sec id="sec-4">
      <title>A usage-based account of the conceptual structure of time</title>
      <p>
        Our goal is to pinpoint the major dimensions of
time in the description of complications in PLMT.
For reasons of concision, we base our depiction
on an often-quoted review article about PLTM
        <xref ref-type="bibr" rid="ref8">(Moreno and Berenguer, 2006)</xref>
        , a common genre
in the medical scholar tradition, that summarizes
available data for a given medical phenomenon.
The article under investigation provides an
overview of allograft dysfunctions and surgical
complications following liver-transplant and discusses
the state-of-the art concerning their medical
follow-up. Strikingly, the article conceptually
opposes immediate and long-term complications.
These complications are of a different nature: they
can be strictly medical (think of respiratory
changes, renal dysfunction or hemodynamic
complications), or technical (complications due to
h(a)emorrhage or vascular complications
resulting, for example, in infections or draft
dysfunctions (major complications in this particular case
are acute cellular rejection or recurrent viral
hepatitis)). Both short and long-term complications
are rather heterogeneous as well, due to the fact
that the liver interacts with very different
subsystems of the organism, whereby any dysfunction
may cause diseases such as chronic rejection,
arterial hypertension, obesity or bone
complications, just to name a few. Our investigation is
limited to the temporal aspects of the different
complications, leaving aside other conceptual
aspects such as anatomical location or severity.
3.1
      </p>
      <sec id="sec-4-1">
        <title>Peri- and post-operative time</title>
        <p>
          A first important time factor is the distinction
between treatment peripheral to the central
(intraoperative) transplant intervention that can have a
direct impact on the reduction of post-surgical
complications
          <xref ref-type="bibr" rid="ref6">(Junttila et al., 2005)</xref>
          , and the time
lapse proper to the complication itself (the
socalled post-operative time). Peri-operative time
has to do with non-problematic follow up of
surgery, before, during and after surgery. The
difference between this peri- and post-operative time is
minimal, given the fact that some complications
such as infections may arise due to improperly
performed medical actions:
(1) The prophylaxis of bacterial infection
includes the following strategies: a) selective
intestinal decontamination; b) administration
of systemic antibiotics peri-operatively, c)
antibiotic prophylaxis before invasive
explorations of the biliary tract, and d) personnel
hand washing together with strict asepsis in
all invasive procedures.
        </p>
        <p>The following observations can be made about
example (1): strategies a), b) and c) in the example
refer to such perioperative precautions. As a part
of medical prevention, peri-operative treatment
contributes to building the temporal barrier ab quo
the time sequence of complication starts to run.
Preventive, pre-operative treatment is an
important time issue that should be taken into
account in modelling medical complications: a
distinction is needed between pre-symptomatic
(example 2) and post-symptomatic treatment of
complications. Prophylaxis aims at non-invasive
avoidance of complication outbreak (example 3)
and therefore reduces medical cost (example 4).
(2) Another form of prevention, mainly targeted
to avoiding the development of clinically
manifest CMV disease, is the treatment of
infection in the pre-symptomatic stage. [81;
note of the authors: CMV refers to
cytomegalovirus, the most frequent micro-organism
in liver transplantation]
(3) Universal prophylaxis is useful mainly in
high-risk patients […] and can be done
effectively and safely with oral drugs […]. [81]
(4) Anticipated treatment is also an effective and
probably most cost-effective strategy.</p>
        <p>The distinction between peri- and post-operative
time as classification parameter is a central issue
in the temporal (when do problems occur?),
causal (what effects are produced?) and
instrumental (what therapy positively affects such
effects?) format of medical decision-making. In
medical discourse, peri-operative techniques are
explicitly juxtaposed to post-operative care, as the
following example shows.
(5) The results of liver transplantation have
improved due to advances in perioperative
technique, a better understanding of the course
and prognosis of several [sic] liver disease
improved immunosuppressive therapy and
more effective postoperative care. [77]
In fact, knowledge about postoperative
complications is actively used in optimal peri-operative
treatment.</p>
        <p>A second time factor at the interface between
peri-operative management and post-operative
complications is the moment of detection of
(early) complication symptoms. Detection and
diagnosis of new symptoms clearly marks the
distinction between prophylaxis and post-operative
care. The following excerpt illustrates this
distinction:
(6) Thus, knowledge of complications that
emerge during follow up period, early and
accurate establishment of diagnosis, and
prompt institution of appropriate
interventions are essential for optimal patient and
graft outcome [77]
An important part of the state-of-the-art consists
in describing so-called early detection methods
such as in the following example:
(7) Methods for early detection of viral
infection, in the case of cytomegalovirus, are
periodic determination of CMV antigenemia in
peripheral blood leukocytes and PCR
techniques to detect the blood viral genome.
Defining medical states is a central issue in liver
transplant surgery. The literature defines the
normal state in the intensive care unit, after
transplantation that shows increasing degrees of recovery
as illustrated in the following example by the
different modified verbal and nominal phrases:
(8) When the transplant evolves favorably, the
patient is awake, hemodynamically stable,
with spontaneous respiration, preserved
renal function, and with progressively
improving liver activity.
3.2</p>
      </sec>
      <sec id="sec-4-2">
        <title>Time of occurrence: immediate and long-term</title>
        <p>Complications are possible alterations to the
desired optimal condition. The most frequent in
PLTM are complications that can be expected
during this early post-transplant period are
hemodynamic alterations, and respiratory, renal and
neurological complications.</p>
        <p>A global distinction is the one between early and
late complications, or more correctly, between
immediate and long-term complications. Late
complications are gaining importance as survival
rates during the early-postoperative period
increases. The distinction between complication
subtypes is based on the point of their occurrence
in time:
(9) The complications occur either immediately
post-transplantation or in the long-term. The
main complications in the immediate
postoperative period are related to the function of
the graft (dysfunction and rejection), the
surgical technique, infections (bacterial, fungal,
and viral), and systemic problems
(pulmonary, renal, or neurological). In the long
term, the complications are typically a
consequence of the prolonged
immunosuppressive therapy, and include diabetes mellitus,
systemic arterial hypertension, de novo
neoplasia, and organ toxicities, particularly
nephrotoxicity.</p>
        <p>An important note is that the underlying
pathology causing the transplantation is not considered
a complication, though the causal trigger may
persist (or reemerge at some later point in time).</p>
        <p>The definition or discursive description of
complications typically contains the designation
of the medical phenomenon, a general
characterization in terms of immediate or late occurrence,
followed by a more precise time label for the time
span within which complications arise (in terms
of hours, as in (10), days, or, in the case of late
complications, months).
(10) A hemorrhage in the immediate
postoperative period is another potential complication
[…].It is typically diagnosed within the first
48 hours post-transplantation (hemorrhagic
abdominal drainages, hemodynamic
instability, serial determination of the
hematocrit/hemoglobin).</p>
        <p>In the above example, the time label is followed
by a summary of medical actions performed
during this time lapse. Modelling complications then
may entail two different time lines: one for
occurrence of complications as such (snap and span,
notions defined in the introduction of this paper),
and one mapping the full (linear or cyclical)
scenario of common medical actions associated with
postoperative care.</p>
        <p>The review article also mentions a shift in the
historical evolution of PLTM complications: as
surgical techniques and immunosuppressive
treatment improve, prognosis and survival chances
increase, correspondingly extending the time span
of (late) medical post-surgical follow-up (cfr. (11)
and (12)).
(11) The main barriers to overcome in the first
period were immediate post-surgical survival
together with prevention of acute rejection.
(12) With greater survival of patients, new
problems have arose that basically affect
transplant recipients with long-term follow up
Long-term complications are rather flexible
conceptual notions. Their emergence is connected
with the specific complication (such as chronic
renal failure, systemic arterial hypertension,
diabetes mellitus, etc.), but at the same time differs
accordingly. Generic time information is
commonly expressed as occurring at a random
moment (13), or by means of unspecified
postoperative time (14).
(13) […] malignant tumors can appear at any time
after transplantation […].
(14) A variable percentage of patients, 4-20%
according to the series, will develop diabetes
mellitus following transplantation (de novo
DM).</p>
        <p>Apart from these general time labels, also
different discursive strategies are used for indicating a
precise moment in time, specifically in late
complications. Here, different scenarios may occur. In
the first one a precise numeric cut-off point after
transplantation is expressed (not until, not before):
(15) Chronic rejection is usually not evident until
at least 6 months after transplant. The
pathogenesis is still unclear.</p>
        <p>In the second scenario, the complication is
associated with a risk decreasing in time, without
mentioning an endpoint:
(16) Arterial hypertension (AHT) is a frequent
complication in liver transplant recipients. Its
prevalence varies between 50-70% in the
first post-transplantation months but
decreases thereafter probably due to the
reduction of the immunosuppressive doses.</p>
        <p>In the third scenario, the full span of time is
expressed (e.g. a one-year period) during which the
complication is most pronounced:
(17) Obesity is a very frequent complication in
transplanted patients […] one year after
transplantation, the period when the greatest
weight gain is seen.
3.3</p>
      </sec>
      <sec id="sec-4-3">
        <title>Timing of occurrence and diagnosis</title>
        <p>An interesting issue is the fact that even within the
subgroup of immediate complications, both the
nature of the complication and its treatment
depend on the moment of discovery: similar sets of
symptoms may cause different kinds of
pathology. This is the case, for example, for the most
frequent complication in pediatric cases:
(18) Symptoms are highly variable and depend on
the timing of development and diagnosis.
(19) When the thrombosis occurs at an early
stage, it typically leads to ischemia/necrosis
of the graft; in contrast, when it occurs at a
later time point, it generally leads to biliary
complications (intrahepatic biliomas and
biliary stenosis) but with preservation of the
graft function.</p>
        <p>This evolutionary, dynamic nature of medical
conditions needs to be taken explicitly into
account in concept modelling of complications. In
its earlier or later diagnostic establishment, a
complication takes up different forms and therefore
requires different treatment types. Consequently, in
the case of thrombosis mentioned in the previous
example, the therapy in the acute or late form
consists of different medical actions:
(20) In the acute form, thrombolysis can be
accomplished by surgical radiology. Arterial
thrombectomy may be an alternative that can
be done either by interventional radiology or
surgical intervention. In patients where these
options fail, urgent re-transplantation may be
required. In the late form, treatment is mainly
focused to prevent/treat biliary
complications derived from the thrombosis.</p>
        <p>Note that in the late form, treatment does not
zoom in on the primary complication, but on the
derived one. This example clearly shows that one
complication may trigger another one. The same
complication may be connected with or caused by
different intentional operations, leading to these
unintended side effects at different stages in
postoperative care:
(21) Biliary fistula can occur initially in the first
month in relation to anastomotic dehiscence
secondary to technical errors or biliary tract
ischemia. It is also a common complication
in the third month when the T-tube is
withdrawn.</p>
        <p>A central issue in this dynamic picture is the
notion of lead time to diagnosis: given the fact that
complications constantly evolve, this time factor
contributes to variable medical decisions, taking
transplantation as a starting point:
(22) The clinical picture is variable and depends
on the time of development, lead time to
diagnosis, and existence of a T-tube.</p>
        <p>We finish this part of our investigation with two
specific subcases of timing. First, some
complications can occur both early, or emerge only late,
after a so-called normal postoperative course. The
following example refers to liver graft
dysfunction:
(23) Dysfunction of the graft may occur in the
immediate postoperative period (early
dysfunction) or late during the follow-up of the
patient {typically related to the recurrence of
the original disease (viral hepatitis, primary
biliary disease, sclerosing cholangitis,
alcohol or autoimmune liver disease) or chronic
rejection}.</p>
        <p>This later complication’s manifestation has a very
distinct causal origin: it is overtly tied to the
original disease, which urged for liver transplantation.
Second, different complications (as the
neurological states) sometimes occur simultaneously: the
first complication is continuous (disorientation),
and then punctuated by episodes of the second one
(agitation and confusion).
(24) The most frequent neurological alterations
are disorientation with episodes of agitation
and confusion.</p>
        <p>A very specific feature of generic late
complications such as malignant tumor is the correlation
between duration (of immunosuppression) and
specific cancer subtypes (Kaposi’s sarcoma, skin
tumors, carcinomas of vulva and perineum):
(25) Although malignant tumors can appear at
any time after transplantation, Kaposi’s
sarcoma followed by lymphoproliferative
disorders are the earliest that usually develop. The
later ones are skin tumors and carcinomas of
the vulva and perineum.
3.4</p>
      </sec>
      <sec id="sec-4-4">
        <title>Duration of complication: rejection from (hyper-) acute to chronic</title>
        <p>Some complications occur either early or late. A
specific case is graft rejection. An interesting time
distinction in this respect is the one between
hyperacute and acute. There is not only a difference
in terms of time itself (hyperacute rejection
occurring within minutes or hours), but also in terms of
the complications’ nature. The prefix hyper-
refers as well to the severity of the rejection reaction
and the fact that antibodies reject the graft in an
irreversible way. Hyperacute rejection is humoral,
whereas acute rejection has a cellular origin. This
is the primary difference with acute rejection, for
which drugs are available. The primary difference
with the third subtype is that
(26) […] Chronic rejection generally occurs over
a span of months, can be unresponsive to
current therapy, and contributes to be a source
of graft loss.</p>
        <p>
          An interesting, yet unsolved issue is the difference
between chronic and repeated acute rejection.
Some studies report that acute rejection generally
occurs in the first few weeks following
transplantation, whereas chronic rejection “typically occurs
several months to a year posttransplantation”
          <xref ref-type="bibr" rid="ref1">(Batts, 1999)</xref>
          and requires additional more
histological fine-tuning by means of liver biopsy.
        </p>
        <p>Determining correct diagnosis for each of
these complications is a highly complex issue,
because of the many clinical parameters shared by
different complications.
3.5</p>
      </sec>
      <sec id="sec-4-5">
        <title>Frequency and prevalence</title>
        <p>
          Frequency and prevalence are commonly used
terms when describing the epidemiological status
of a complication
          <xref ref-type="bibr" rid="ref5">(Greenberg et al., 2005, chapter
2)</xref>
          . Whereas incidence refers to the number of new
cases occurring in a given period of time,
prevalence indicates the actual number of cases alive
either during a period or at some point in time.
        </p>
        <p>The prevalence can be addressed in different
ways: on average and based on variation (either
within the different complications of a subtype or
as a sample within the complication population).
The following three examples illustrate this
characterization in three progressive steps.
(27) The prevalence of technical complications is
on average 26%.
(28) Arterial complications, particularly the
thrombosis of the hepatic artery (prevalence
ranging from 1.5 to 25%) are the most
frequent ones.
(29) Hepatic artery thrombosis is a complication
that develops more frequently in the pediatric
population.</p>
        <p>Apart from overall values (27-28) for the
population, samples typically address age, such as
children (29), specific pre-surgical diseases triggering
grafting (30), and retransplant patients (31):
(30) Portal vein thrombosis is an infrequent
complication with an overall prevalence of 2-3%.
(31) Globally, 20-40% of liver transplant
recipients present atraumatic bone fractures; this
prevalence rises to 65% in patients
transplanted due to cholestatic disease and in
retransplant patients.</p>
        <p>Typical long-term spans include mention of
one and five year periods.
(32) Currently though, survival rates of over
9095% and 70% at one year and five years
posttransplantation, respectively are expected.</p>
        <p>Prevalence spans reach from 100% to values
as close as 1.5%, as in (33) and (28).
(33) Pleural leakage, predominantly on the right,
is the most frequent complication with a
prevalence reported to be as high as 100% in
some series.</p>
        <p>Different diagnostic criteria are used to
characterize a (chronic) complication. Variable prevalence
is therefore a common measure:
(34) The prevalence is variable, depending on the
criterion used to define it and to the method
used to assess renal function. Indeed, serum
creatinine measurement may underestimate
the presence of renal failure.</p>
        <p>Particularly in late complications, an
important comment on the relation between time and
treatment has to be made: a complication may
display decreasing prevalence over time. As
mentioned before, complications are sometimes
caused by immunosuppressive drugs. Since drug
administration may be decreased over time, this
has an impact on its frequency. Conceptually,
there is again a correlation (or even causal
relation) between treatment method and time,
independent of a specific complication, as the two
following examples (concerning hypertension and
diabetes, respectively) show.
(35) Arterial hypertension (AHT) is a frequent
complication in liver transplant recipients. Its
prevalence varies between 50-70% in the
first post-transplantation months but
decreases thereafter probably due to the
reduction of the immunosuppressive doses.
(36) A variable percentage of patients, 4-20%
according to the series, will develop diabetes
mellitus following transplantation (de novo
DM). The prevalence depends on the time
elapsed since transplantation and particularly
on the immunosuppressive drugs. In the
initial post-transplantation period, DM is very
frequent, probably due to the use of high CNI
and steroid doses.</p>
        <p>Typically, measures for prevalence and risk
may compare patients with complications to the
healthy population (37), taking into account the
moment of grafting (38) and stage of drug
development.
(37) The natural history of malignant tumors in
the transplant patient tends to be different
from that of the normal population; they
appear at an earlier age, tend to be in a more
advanced stage when diagnosed, and their
evolution is more aggressive, causing high
mortality directly related to the tumor.
(38) Some data suggest that in patients
undergoing liver transplantation in recent years, there
is a higher incidence of hematological
neoplasms with de novo internal neoplasms
developing at earlier time-points than in those
transplanted years ago.</p>
        <p>A relevant measure and objective of PLTM
management is the reduction of frequency of
complications to values for the general population, as in
the case of bacterial infections:
(39) After the sixth month, with the transplanted
organ functioning normally and minimum
immunosuppressive doses, the frequency of
bacterial infections is reduced to figures
similar to those of the general population and the
causes are pathogenic bacteria of the
community.</p>
        <p>Liver transplantation entails serious risks.
Prevalence therefore is not only coined in terms of
morbidity, but also of (decreasing) global mortality.
(40) The global mortality in this early
posttransplantation period is approximately 5-10%.
Particularly in retransplantation, mortality rises.
Retransplantation entails two subsequent time
sequences: the management of the first graft,
leading to an incurable complication (such as graft
rejection mentioned before) within the first 48
hours following surgery, and a second one, with
reduced survival prognosis.
(41) However, if regression of the clinical
situation is not observed after 24-48 hours,
retransplantation must be considered as soon as
possible to avoid the development of
multiorgan failure, in which case the mortality
associated with retransplantation is very high.
4</p>
      </sec>
    </sec>
    <sec id="sec-5">
      <title>Conclusion</title>
      <p>This paper proposes a first sketch of time factors
as an associative relation relevant for modelling
the temporal unfolding of PLTM complications.
We list the time factors useful for a model of time.
1. A boundary separates peri-operative care and
management of complications. Prophylaxis is</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          <string-name>
            <surname>Batts</surname>
            ,
            <given-names>K.P.</given-names>
          </string-name>
          <year>1999</year>
          .
          <article-title>Acute and chronic hepatic allograft rejection: pathology and classification</article-title>
          .
          <source>Liver Transplantation and Surgery</source>
          <volume>5</volume>
          (
          <issue>44 Suppl 1</issue>
          ):
          <fpage>S21</fpage>
          -
          <lpage>9</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          <string-name>
            <surname>Clavien</surname>
          </string-name>
          ,
          <string-name>
            <surname>Pierre</surname>
          </string-name>
          .-Alain, Carlos A.
          <string-name>
            <surname>Camargo</surname>
            , Ruth Croxford, Bernard Langer,
            <given-names>Gary A.</given-names>
          </string-name>
          <string-name>
            <surname>Levy</surname>
            &amp;
            <given-names>Paul D.</given-names>
          </string-name>
          <string-name>
            <surname>Greig</surname>
          </string-name>
          .
          <year>2006</year>
          .
          <article-title>Definition and classification of negative outcomes in solid organ transplantation. Application in liver transplantation</article-title>
          .
          <source>Annals of Surgery</source>
          <volume>220</volume>
          (
          <issue>2</issue>
          ):
          <fpage>109</fpage>
          -
          <lpage>120</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          <string-name>
            <surname>Dindo</surname>
          </string-name>
          , Daniel, Nicolas Demartines &amp;
          <string-name>
            <surname>Pierre-Alain Clavien</surname>
          </string-name>
          .
          <year>2004</year>
          .
          <article-title>Classification of Surgical Complications. A New Proposal with Evaluation in a Cohort of 6336 Patients and Results of a Survey</article-title>
          .
          <source>Annals of Surgery</source>
          <volume>240</volume>
          (
          <issue>2</issue>
          ):
          <fpage>205</fpage>
          -
          <lpage>213</lpage>
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          <string-name>
            <surname>Grabar</surname>
            ,
            <given-names>N.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Hamon</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Bodenreider</surname>
            ,
            <given-names>O.</given-names>
          </string-name>
          <year>2012</year>
          .
          <article-title>Ontologies and terminologies: continuum or dichotomy?</article-title>
          <source>Applied Ontology</source>
          <volume>7</volume>
          (
          <issue>4</issue>
          ):
          <fpage>375</fpage>
          -
          <lpage>386</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          <string-name>
            <surname>Greenberg</surname>
            , Raymond,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Stephen</surname>
            <given-names>R.</given-names>
          </string-name>
          <string-name>
            <surname>Daniels</surname>
            ,
            <given-names>W.</given-names>
          </string-name>
          <string-name>
            <surname>Dana</surname>
            <given-names>Flanders</given-names>
          </string-name>
          , John William Eley &amp;
          <string-name>
            <surname>John R. Boring</surname>
          </string-name>
          <article-title>III (eds</article-title>
          .).
          <year>2005</year>
          .
          <string-name>
            <given-names>Medical</given-names>
            <surname>Epidemiology</surname>
          </string-name>
          . New York:
          <string-name>
            <surname>McGraw-Hill</surname>
          </string-name>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          <string-name>
            <surname>Junttila</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Salantera &amp; M. Hupli</surname>
          </string-name>
          .
          <year>2005</year>
          .
          <article-title>Developing terminology for documenting perioperative nursing interventions</article-title>
          .
          <source>International Journal of Medical Informatics</source>
          .
          <volume>74</volume>
          (
          <issue>6</issue>
          ):
          <fpage>461</fpage>
          -
          <lpage>471</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          <string-name>
            <surname>Khoo</surname>
            ,
            <given-names>C. &amp; N.</given-names>
          </string-name>
          <string-name>
            <surname>Jin-Cheon</surname>
          </string-name>
          .
          <year>2006</year>
          .
          <article-title>Semantic Relations in Information Science</article-title>
          .
          <source>Annual Review of Information Science and Technology</source>
          <volume>40</volume>
          :
          <fpage>157</fpage>
          -
          <lpage>228</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          <string-name>
            <surname>Moreno</surname>
            ,
            <given-names>R.</given-names>
          </string-name>
          &amp;
          <string-name>
            <surname>M. Berenguer</surname>
          </string-name>
          .
          <year>2006</year>
          .
          <article-title>Post-liver transplantation medical complications</article-title>
          .
          <source>Annals of Hepatology</source>
          <volume>5</volume>
          (
          <issue>2</issue>
          ):
          <fpage>77</fpage>
          -
          <lpage>85</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          <string-name>
            <surname>Munn</surname>
            ,
            <given-names>Katherine &amp; Barry</given-names>
          </string-name>
          <string-name>
            <surname>Smith</surname>
          </string-name>
          .
          <year>2008</year>
          .
          <article-title>Applied ontology</article-title>
          .
          <source>An Introduction. Frankfurt</source>
          and Paris: Ontos.
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          <string-name>
            <surname>Pareja</surname>
            ,
            <given-names>E.</given-names>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Cortes</surname>
          </string-name>
          ,
          <string-name>
            <given-names>R.</given-names>
            <surname>Navarro</surname>
          </string-name>
          ,
          <string-name>
            <given-names>F.</given-names>
            <surname>Sanjuan</surname>
          </string-name>
          ,
          <string-name>
            <given-names>R.</given-names>
            <surname>López</surname>
          </string-name>
          &amp;
          <string-name>
            <surname>J. Mir</surname>
          </string-name>
          .
          <year>2010</year>
          .
          <article-title>Vascular complications after orthotopic liver transplantation: hepatic artery thrombosis</article-title>
          .
          <source>Transplantation Proceedings</source>
          <volume>42</volume>
          :
          <fpage>2970</fpage>
          -
          <lpage>2972</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          <string-name>
            <surname>Roche</surname>
          </string-name>
          , Christophe.
          <year>2007</year>
          .
          <article-title>Terme et concept: fondements pour une ontoterminologie</article-title>
          .
          <source>Proceedings TOTh 2007 Terminologie &amp; Ontologie: Théories et Applications. Annecy</source>
          ,
          <volume>1</volume>
          -
          <fpage>22</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          <string-name>
            <surname>Sambre</surname>
            ,
            <given-names>Paul &amp; Cornelia</given-names>
          </string-name>
          <string-name>
            <surname>Wermuth</surname>
          </string-name>
          .
          <year>2010</year>
          .
          <article-title>Causal framing for medical instrumentality: applied ontology and frame-based construction grammar</article-title>
          .
          <source>Belgian Journal of Linguistics</source>
          <volume>24</volume>
          :
          <fpage>163</fpage>
          -
          <lpage>191</lpage>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>